3".“
fig;

*2

.. ‘
r‘v .
as,

up. A
.A J“ .
fig...“
, a?
c

u

..

7‘ A...,

nah

r4

£33, in nan .. ‘
gr»: 4%“?

.
‘i-
:2
S? r 3.... J. ,

\

.
I.

. “an”

$13.1 «1.... if. v.
. t. 9 i. .\
. K'IKDWA 3.4. 59¢ yr.
! :flhar ..
,x y‘

no
a.
«i

 

 

 

I...

 

 

2:07
This is to certify that the
dissertation entitled
Parent Perceptions of the Home School Relationship:
Understanding the Model, the Psychological Mechanism, and
Moderating Factors
presented by
Jana L. Aupperlee
l l
o
p. E >, has been accepted towards fulfillment
g In a of the requirements for the
G a
m "its; Doctoral degree in Counseling, Educational
:3 .n c Psychology, and Special
.2 9 Education
2
L_____.J

    

x" l 7
t/ ‘ k/

j Major Professor’s Signature

May 10, 2007

 

Date

MSU is an affirmative-action, equal-opportunity employer

..-n---n—l_.-l-I-I-I'D-I-l-O-l-l—O-l-l-

In—A-I-I-c-I-l-I-o-—l-o-O-I-l-l-‘-I--l--l-h-l-o-I- -

 

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

 

DATE DUE DATE DUE DATE DUE

$301291”

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

6/07 p:/CIRC/DateDue.indd-p.1

PARENT PERCEPTIONS OF THE HOME-SCHOOL RELATIONSHIP:
UNDERSTANDING THE MODEL, THE PSYCHOLOGICAL MECHANISM, AND
MODERATING FACTORS
By

Jana L. Aupperlee

A DISSERTATION
Submitted to
Michigan State University

in partial fulfillment of the requirements
for the degree of

DOCTOR OF PHILOSOPHY
Department of Counseling, Educational Psychology, and Special Education

2007

ABSTRACT
PARENT PERCEPTIONS OF THE HOME-SCHOOL RELATIONSHIP:
UNDERSTANDING THE MODEL, THE PSYCHOLOGICAL MECHANISM AND
MODERATING FACTORS
By
Jana L. Aupperlee
Parents and schools are the primary socializing agents in the lives of children and
as such, the relationship between the two groups is vitally important. This study
unraveled how school-provided support influenced school-related outcomes for the
parents of elementary school students. Study results revealed a number of findings.
First, this study suggested that the home-school relationship might consist of an affective
and a behavioral component contrary to theory in the field (e.g., Esler, Godber, and
Christenson, 2002). In addition, this study offered evidence for social support as the
psychological mechanism by which the home-school relationship operates; school-
provided social support accounted for a sizeable amount of variance in parent trust and
satisfaction and small amount of variance in parent involvement. This study also
suggested that school-provided social support consists of a single factor as opposed to the
emotional and instrumental structure espoused in previous research. Finally, this study
showed that school location and parent education changed the relationship between social
support and parent satisfaction. In particular, suburban parents indicated lower
satisfaction than urban parents with lower levels of social support and higher satisfaction

with more social support. Parent education also changed the relationship. Parents with

less education indicated lower satisfaction than parents with more education with less

support and higher satisfaction with more support. Affective and behavioral strategies

for transforming the home-school relationship were presented.

Copyright by
JANA L. AUPPERLEE
2007

DEDICATION

I dedicate this text and this degree to my family whose unending love and support made it
possible.

Mark and Andrew: thank you for your unceasing support throughout this process. Mark,
you helped me to appreciate the humor inherent in academic pursuits. Drew, you put my
work in perspective.

John, Lynn, and Christina Gezon, and William, Susan, and Todd Aupperlee: Thank you
for your phone calls, e-mails, thoughts, and prayers throughout my academic career. You
consistently promoted my mental health.

ACKNOWLEDGEMENTS

Completing this dissertation was a joy. While I began the process with
trepidation, I would not change a single step of this project. The completion of my
dissertation has shaped my professional identity, illustrating the integration of research
and practice, and helping me to feel like a full-fledged member of the academic
community. This process would not have been as instrumental in my development
without the support Of the faculty and my colleagues in the School Psychology program
at Michigan State University.

First, I would like to acknowledge my committee members, Jean Baker (director),
Evelyn Oka, John Carlson, and Joanne Keith. To Jean, thank you for being an unflagging
cheerleader and for helping me to write a “good enough” dissertation. To Drs. Oka,
Carlson, and Keith, thank you for your thoughtful and timely feedback. You’re helped

me to create a better text and you’ve inspired me to think like a scholar.

vi

TABLE OF CONTENTS

LIST OF TABLES ........................................................................................................... viii
LIST OF FIGURES ........................................................................................................... ix
Chapter One: Introduction .................................................................................................. 1
Chapter Two: Review of the Literature ............................................................................ 10
Home-School Relationships .......................................................................................... 1 1
Social Support ............................................................................................................... 28
Moderation: Introduction and Specific Moderators ...................................................... 50
Research Questions and Hypotheses ............................................................................ 60
Chapter Three: Methodology ............................................................................................ 63
The General Perspective ............................................................................................... 63
Procedures ..................................................................................................................... 69
Measures Used in Data Collection ................................................................................ 73
Chapter Four: Results ....................................................................................................... 80
Measure Preparation ..................................................................................................... 80
Examination of Hypotheses .......................................................................................... 90
Chapter Five: Discussion ................................................................................................ 103
Social Support as a Psychological Mechanism ........................................................... 103
Examination of a Theoretical Model of Home-School Relationships ........................ 109
Moderators of the Home-School Relationship ........................................................... 110
Implications of the Study as a Whole: Affective and Behavioral Strategies for
Improving the Home-School Relationship ................................................................. 1 14
Limitations and Additional Research .......................................................................... 117
Future Directions ........................................................................................................ 118
Appendix One: Teacher Letter ........................................................................................ 122
Appendix Two: Parent Letter .......................................................................................... 123
Appendix Three: Home-School Relationship Survey ..................................................... 124
Appendix Four: Parent Consent ...................................................................................... 130
Appendix Five: Reminder Letter .................................................................................... 132
References ....................................................................................................................... 133

vii

LIST OF TABLES

Table 1: Demographic Information for Participating Schools .......................................... 65
Table 2: Demographic Information for Participating Parents ........................................... 68
Table 3: Summary of Reliability for Existing Measures ..................................... 82

Table 4: Descriptive Statistics for Social Support, Parent Outcomes, and Moderators
Prior to Transformation ............................................................................................. 84

Table 5: Descriptive Statistics for Social Support, Parent Outcomes, and Moderators
After Transformation ................................................................................................ 85

Table 6: Correlations Between Measures ......................................................................... 87

Table 7: Summary of Hierarchical Regression Analysis for Social Support and Child
Grade in Predicting Parent Trust, Parent Satisfaction, and Parent Involvement ...... 92

Table 8: Summary of Hierarchical Regression Analysis for Social Support and Parent
Education in Predicting Parent Trust, Parent Satisfaction, and Parent Involvement 96

Table 9: Summary of Hierarchical Regression Analysis for Social Support and School
Location in Predicting Parent Trust, Parent Satisfaction, and Parent Involvement 100

viii

LIST OF FIGURES

Figure 1: Wave Analysis Results for Parent Responses Over Time ................................. 73
Figure 2: Social Support and Parent Satisfaction as Moderated by Parent Education ..... 97

Figure 3: Social Support and Parent Satisfaction as Moderated by School Location 101

Chapter One: Introduction

With the rise in accountability and growing concern over failing schools and
children, home—school collaboration is a hot topic in education. A variety of programs,
including Title One, the National Educational Goals Panel, the National Parent Teacher
Association, the United Stated Department of Education, and numerous other
organizations have brought the relationship between families and schools to the forefront
of the national consciousness (Esler, Godber, & Christenson, 2002). As a potential
solution, federal and state funds have been allocated to promoting and enhancing parent
involvement in schools.

Despite this financial and policy-related investment in home-school collaboration,
research clearly Show that children are failing to meet basic standards and that parent
satisfaction with school is on the decline. For instance, data from the National Center for
Education Statistics (NCES; Wirt et al., 2004), indicated that in 2002, only 31 percent of
fourth graders, 33 percent of eighth graders, and 36 percent of twelfth graders were
reading at or above the proficient level, a key indicator of academic success. Further,
NCES (2002) research reveals that the percentage of parents of third through twelfth
graders reporting being “very satisfied” with their children’s schools decreased fi'om 56
percent in 1993 to 53 percent in 1999.

Despite this gn'm student achievement and parent attitudinal data, national data
also offers some hope about the state of education, particularly in regards to parent
involvement at school and family support of education at home. In particular, the
percentage of parents who attended a general school meeting during the 1996 and 2002

school years remained consistent at 88 percent (NCES, I999; Vaden-Kiernan &

McManus, 2005). Further, other markers of parent involvement showed increases during
that time span. Specifically, the percentage of parents attending a school event increased
from 66 to 70 percent and the percentage of parents who acted as a volunteer or served on
a school committee increased from 40 to 42 percent (Vaden-Kieman & McManus, 2005).
Parent support of education at home also increased. Specifically, “the percentage of
children who were read to by a family member three or more times per week increased
from 78 percent in 1993 to 84 percent in 2001” WCES, 2003, p. 75). In addition, the
percentage of children whose family members taught them letters, words, or numbers
increased from 58 to 74 percent (NCES, 2003).

Taken together, these data suggest that academic achievement could improve and
that parent satisfaction with school is waning. On the other hand, evidence also indicates
that parents are becoming more involved at school and in supporting learning at home.
Brief Introduction to the Home-School Relationship Literature

Despite the aforementioned statistics on student achievement, parent satisfaction,
and parent involvement, the relationship between families and schools is at best poorly
understood. This field lacks clarity for a number of reasons. First, the literature
supporting home-school collaboration is overridden with competing models and
constructs, very few of which have been empirically examined. Second, very few studies
have proposed a psychological mechanism by which this phenomenon operates, which
could be related to the overabundance of theories. In addition, while research clearly
indicates that parents make a difference in educational outcomes for children, parent
outcomes associated with their relationship with the school have been largely ignored.

Finally, researchers have yet to examine how moderators like school location and parent

educational level change the nature of the relationship between home-school
collaboration and various parent outcomes.
Key Definitions

One way to begin to impose order on this field is for researchers to clearly define
important terms. Thus, a brief introduction to the key terms in this study follows below.
The home-school relationship is the focus of this study. Briefly, the home-school
relationship is defined as parent and school perceptions and practices that characterize the
relationship between the adults who care for a child in the home context and the adults
who contribute to the learning of the child in the school context.

Social support is the hypothesized mechanism by which the family-school
relationship operates. In this study, social support will be defined after Cauce, Felner and
Primavera’s 1982 model as “the range of significant interpersonal relationships that have
an impact on an individual’s functioning” (p. 418). Although this definition neatly
highlights human relational needs and the active role played by social support, the
definition is too narrow. Thus, the definition will be broadened to include the support
provided by organizations or institutions such as schools (House, 1981).

Three variables pertaining to school comprise this study’s outcomes. The first
outcome is parent trust, which will be defined according to Adams and Christenson’s
(2000) model as “confidence that another person will act in a way to benefit or sustain the
relationship, or the implicit or explicit goals of the relationship, to achieve positive
outcomes for students” (p. 480). Second, parent satisfaction with school will be included,
which is a family’s affective appraisal of their child’s school life and experiences.

Finally, parent involvement will be defined somewhat narrowly as parent-initiated

involvement activities focused on their children’s school (Kohl, Len gua, & McMahon,
2000)
Issues Surrounding the Home-School Relationship

As noted previously, numerous studies have attempted to unravel the theoretical
tangle of home-school collaboration. In particular, Epstein (1995) proposed a six-
component typology of behaviors that comprise home/school/community partnerships.
Her model included a variety of roles and tasks for families including parenting,
communicating between the home and school, volunteering at school, fostering learning
at home, sharing responsibility for decision making in the school, and collaborating with
the community (Epstein, 1995; Epstein & Dauber, 1991).

Esler, Godber and Christenson (2002) have also attempted to bring order to this
unruly field. Their model incorporates the actions, relationships, and attitudes involved
in home-school collaboration. Although widely disseminated, this theory has at best
meager empirical backing.

Although Epstein (1995) and Esler, Godber and Christenson (2002) present
plausible, clear models, neither model has the scientific rigor necessary for advancing our
knowledge in this area. In order for further growth in this field, I empirically examined
Esler, Godber, and Christenson’s model of the home-school relationship.

In addition, researchers have begun to tease apart different components of the
home-school relationship. For instance, predictors of involvement like race (e.g.,
Marcon, 1999) and income level (e.g., Desimone, 1999) have been clearly documented,
as have outcomes associated with home-school collaboration like student academic

achievement (i.e., Fan & Chen, 2001). Despite clarity around the periphery of the

relationship, scholars studying home-school collaboration have largely failed to specify a
psychological mechanism by which the relationship Operates. In other words, they have
failed to specify exactly what comprises the relationship between families and schools.

One notable exception to the dearth of psychological mechanisms comes from
Grolnick and Slowiaczek (1994). They proposed and examined a motivational model by
which parent involvement in schools affected child grades. In particular, they
hypothesized that components of parent involvement influenced child variables including
self-regulation, perceived competence, and control understanding, which in turn, affected
child grades. In testing their model, Grolnick and Slowiaczek (1994) found that
perceived competence and control understanding significantly influenced child grades.
While their model is conceptually and empirically clear, it has been largely forgotten as
scholars rush to understand parent involvement.

Another potential mechanism to explain this relationship, and one that is unique
within the field comes from the realm of psychology rather than education. Specifically,
the construct of social support has the potential to change the way scholars conceptualize
the home-school relationship. As noted before, social support is the range of
relationships with individuals and organizations that influence an individual’s
functioning. This definition calls attention to the relational aspects of social support.
Further, it highlights the internal or phenomenological nature of the construct; perceived
social support rests within an individual person. In this study, social support is
conceptualized to flow from schools as systems to parents and families.

In addition to not specifying a psychological mechanism, a further difficulty

within the field is that researchers have almost exclusively examined child outcomes

associated with the home-school relationship. Although child outcomes like academic
achievement are indubitably important (i.e., Pettit, Bates, & Dodge, 1997), they only
comprise one part of the home-school relationship. Parent outcomes deserve study as
well, particularly since parents are the most proximal figures in the relationship.

A final gap within the field is that research has not established how the
relationship between the home and schools changes in the presence of moderators
including child grade, parent race, parent educational attainment, and school location. In
particular, regarding child grade level, research suggests that the parents of children in
earlier grades are more involved than the parents of Older children (lzzo, Weissberg,
Kasprow, & Fendrich, 1999). One purpose of this study is to ascertain whether this
remains true when the mechanism of social support is used to explain home-school
collaboration and when parent outcomes are examined rather than child outcomes.

A second moderator that has the potential to change the relationship between
home-school collaboration and parent outcomes is parent race. A long tradition of
literature has demonstrated that race influences parent involvement activities (i.e.,
Fehrmann, Keith & Reimers, 1987; F euerstein, 2000). This study examines whether their
relationship with school influences outcomes for parents of different racial and ethnic
backgrounds.

Another potential moderator in the relationship between home-school
collaboration and parent outcomes is parent educational attainment. Research has long
demonstrated that parents without post-secondary education have lower levels of parent
involvement in the schools, at least when involvement is traditionally defined

(Bempechat, 1992). With a broader definition of involvement including support of

school at home, the relationship deserves further study (Grolnick & Slowiaczek, 1994).
One aim of this study is to examine the link between educational attainment and parent
outcomes, rather than child outcomes, as the dependent variable.

The final potential moderator of this relationship is school location. Although it is
intuitive that school locations ranging from suburban to urban would influence outcomes
of the home-school relationship, little research has examined this association. In
particular, Thompson (2002) examined Detroit schools and found that location
significantly predicts child academic achievement. However, the influence exerted by
school location on parent outcomes has not been examined empirically.

Importance for School Psychology

The relationship between families and schools holds great potential for the field of
school psychology. First, school psychologists are uniquely positioned to act as change
agents and to facilitate relationship development. Second, as scientist-practitioners
invested in the production, consumption, and evaluation of research, psychologists have a
responsibility to further the theory supporting practices in schools. Third, better
understanding the relationship between families and schools provides a solid base for
designing interventions to improve the relationship.

In straddling the world of school and the world of families, school psychologists
are uniquely positioned to act as change agents in school systems (Esler et al., 2002).
Because school psychologists are a part of the school community yet work closely with
parents and students, they have the advantage Of multiple perspectives and being able to
work on all sides to improve the relationship between families and schools. In addition,

school psychologists have unique knowledge about systemic functioning and

communication. Specifically, with problem-solving Skills and the ability to engage in
systematic data collection, school psychologists have real power to change the standard
Operating procedure in schools. Finally, due to their role, school psychologists have the
power to change or at least influence school- and district-wide policies related to
collaboration and thus have the responsibility to facilitate the relationship between
parents and schools (Demaray & Malecki, 2002).

In addition to being uniquely positioned to facilitate change in schools, school
psychologists should also be invested in understanding the home-school relationship
because of close ties with the world of scholarship. Psychologists who adopt a scientist-
practitioner orientation act as consumers, producers, and evaluators of research (Hayes,
Barlow, & Nelson-Gray, 1999). A commitment to being a scientist-practitioner entails a
number of responsibilities pertaining to home-school collaboration. First, school
psychologists have a responsibility to understand and develop the literature supporting
this construct. As noted elsewhere, the empirical work supporting this construct is still in
its infancy. In addition, as producers of research, school psychologists Should strive to
document, evaluate, and be accountable for their activities pertaining to home-school
collaboration. For psychologists practicing in schools, this means that systematic data
collection pertaining to the home-school relationship is necessary. Finally, as evaluators
of research, school psychologists have a responsibility to critically review and follow best
practices in enhancing home-school collaboration.

Third, better understanding home-school collaboration is important for school
psychology because of the fields’ commitment to using interventions to improve

education for children and families (Ysseldyke et al., 1997). As Griffith (1997) notes,

changing school attitudes and climate can create real change in family satisfaction with
schools. Further, since clear evidence suggests that parents are important in the
educational lives of children (Henderson & Berla, 1997), and that the relationship with
the school influences parents (e.g., Griffith, 1997), school psychologists have a
responsibility to bring parents into the educational process in a practical and lasting way.
This is particularly important since gaining knowledge about what type of support
families need is the only way to design effective interventions (Cutrona & Russell, 1990).
Until psychologists have established how this relationship works, what comprises the
outcomes, and how it differs for families based on demographic information like child
grade and race, school psychologists will not be able to design interventions that improve
the relationship for all stakeholders.

In summary, the purpose of this study is to examine a theoretical model of the
home-school relationship. In particular, this study will explore how parent perceptions of
the home—school relationship affect parent outcomes related to their children’s schools.
The home-school relationship is thought to operate through the psychological mechanism
of social support as defined by House (1981). In accordance with the model proposed by
Esler et al. (2002), home-school collaboration, through the mechanism of social support,
is hypothesized to influence attitudes, relationships and actions. Parent outcomes
corresponding with these three components include: parent satisfaction with school
(attitudes), parent trust of the school (relationships), and parent involvement with the
school (actions). The second part of the study examines whether this relationship
between home-school collaboration and parent outcomes differs by child grade, parent

race, parent educational attainment, and school location.

Chapter Two: Review of the Literature

Families comprise the primary developmental context and socializing force in the
lives of children (Scott-Jones, I995). The second most important context in which
children develop and learn is the schools (Epstein, 1995). As Bronfenbrenner (1977) and
others have noted, the relationship between the systems in which children are raised is
vitally important (Christenson & Buerkle, I999; Grolnick & Slowiaczek, I994). Luthar,
Cicchetti, and Becker (2000) note that from an ecological perspective, these “levels
transact with each other over time in shaping ontogenic development and adaptation” (p.
552)

For decades, researchers in education and psychology have attempted to
understand how the family and school contexts work together to support the
developmental and educational needs of children (e.g., Ames, 1993). Terms used to
address this relationship have varied widely, including “parent involvement, family-
school partnerships, family-support and family-centered practices” (Christenson &
Godber, 2001 , p. 455). Most recently, this literature has existed under the heading of the
home-school relationship (Christenson & Godber; Esler et al., 2002). Regardless of the
term, the relationship specified and examined is that between the home and school
contexts of children. This is separate from the parent involvement literature that focuses
on school-wide, systems-level initiatives designed to engage parents in the educational
process. Examples of these parent involvement initiatives include bringing parents in as
decision-makers and developing parent teacher organizations (e.g., Hara & Burke, 1998;

Marcon, I999).

10

Home-School Relationships

For this study, the interaction between families and schools will be termed the
home-school relationship. The home component of this phrase suggests that a child’s
context of origin is not limited to parents or even to a family, which takes a variety of
forms in our increasingly diverse society. Home refers to the child’s developmental
setting and all the people who contribute to the growth of the child and to the individuals
who comprise the context, rather than the physical space in which they reside.

The second component of the term is school, which refers primarily to the people
within the education system who influence the lives of children. This term is
purposefully broad as a corrective against more narrow definitions of school (Bempechat,
1992). The term school incorporates all staff members such as school psychologists,
administrators, teachers, and support staff as well as the students. In addition, the temr
includes the policies and practices of the school. For instance, a school’s mission
statement or motto like “be safe, be respectful, and be responsible” or a school- or
classroom-wide policy of teachers calling parents when their children are missing
homework assignments could both be included in a definition of school. Again, the
definition refers not to the building but to the people, policies, and practices subsumed
under the heading of school.

The third part of the term is relationships. Again, relationships are broadly
defined so that a variety of parenting and schooling practices can be included.
Relationships refer to interactions between individuals and systems. These interactions
include both family and school perceptions of one another and their relationship and

observable behaviors. For instance, relationships could include everything from parent

ll

perceptions Of school warmth to more traditional components like teachers sending home
notes and parents volunteering in classrooms. From an ecological perspective, these
“interrelations among major settings containing the developing person at a particular
point in his or her life” are crucial to fully understanding child development
(Bronfenbrenner, 1977, p. 515). Briefly, the home-school relationship is comprised of
parent and school perceptions and practices that characterize the relationship between the
adults who care for a child in the home context and the adults who contribute to the
learning of the student in the school context.
Models of Home-School Relationships

Although literature supports the idea that home-school relationships are essential
to development (Epstein, 1995; Pettit et al., 1997), the specific model by which it
operates is less well understood. Researchers within various fields have posited models
explaining the relationship. Three influential and oft-cited models include those
presented by Joyce Epstein, Diane Scott-Jones, and Sandra Christenson and her
colleagues. Although the Scott-Jones model is not a specific model ofhome-school
relationships, it still presents a comprehensive model of one facet of parent engagement
in education.
Epstein’s Model

Joyce Epstein (1995) writes from within the field of school psychology. She
examined school/family/community partnerships in schools and presented a theory
designed to foster the success of all children “in school and in later life (p. 701).” Her
theory is based on the idea that the three spheres of influence in the lives of children

(families, schools, and communities) exist in relationship to one another. Specifically,

12

Epstein notes that these spheres should work together to create school-like families and
family-like schools where lcaming is promoted and development is optimized. She
breaks the relationship down into six essential components including: parenting,
communicating between the home and school, family members volunteering at school,
fostering learning at home, shared responsibility for decision making in the school, and
collaborating with the community (Epstein & Dauber, 1991). In order to enact these
components, Epstein specifies a number of school-based behaviors including creating an
action team, obtaining funds and other support, identifying starting points, developing a
three-year plan, and continuing to plan and work together.

Scott-Jones’ Model

Diane Scott-Jones (I995) explores the family component of the home—school
relationship. She conceptualizes how parents engage in academic socialization and
beyond that, how they actively participate in assisting children with schoolwork. In
particular, Scott-J ones postulates a four level model. First, parents value specific subjects
and education as a whole. In particular, parents value specific subjects like reading and
mathematics. Parents’ valuing of education includes educational expectations, valuing
effort, acting as models, and providing educational resources.

The second component of the Scott-J ones model is “monitoring school
performance and activities that enhance or diminish school performance” (p. 87).
Positive monitoring activities include setting rules and monitoring activities such as
television viewing (Fehrrnann et al., 1987).

Another component of the model is helping the child complete work. This

component incorporates families fostering basic academic skills through the expert—

l3

novice relationship elucidated by Vygotsky (e. g., Boekaerts, 1999) and parents assisting
children with homework.

The fourth component, which has negative connotations, is doing, which refers to
families becoming over-engaged in completing work assigned to their children. Doing
occurs when parents take responsibility for assignments and attach consequences to task
completion rather than supporting their children’s learning, whereby reducing intrinsic
motivation (Gottfried, Fleming, & Gottfiied, 1994).

E51 er. Godber and Christenson’s Model of Home-School Collaboration

Esler, Godber and Christenson (2002) offer the most comprehensive model of
home-school relationships. Their model is unique because it accounts for attitudes and
relationships as well as the actions that comprise traditional parent involvement. They
hypothesize that these three components are interconnected, influence one another, and
are all necessary for optimal home-school relationships (Esler et al., 2002).

Attitudes form the basis of the model. Specifically, Esler and colleagues (2002) describe
school—wide attitudes towards collaboration as the primary factor in determining whether
families are involved in their children’s schools. They cite Epstein and Dauber’s 1991
research as proof. It showed that schools reaching out to parents was more important
than parent demographic variables in determining levels of parent involvement. Esler
and colleagues also describe the evidence-based components that comprise successful
attitudes. First, shared responsibility between the home and the school, or a “posture of
reciprocity” is essential. This posture of reciprocity is defined as the school providing
official channels for reciprocal discourse between homes and schools (Esler et al., 2002).

Schools also need to view parents as the experts on their children. Third, should adopt a

14

preventive, solution-oriented focus (Esler et al., 2002). This approach is characterized by
relationships that are non-blaming and foster consistent messages across settings.

The second major component of the Esler et al. (2002) model is the relationships
that maintain and support the partnerships. Esler and her colleagues present a list of Six
practices schools and particularly school psychologists should encourage to promote
healthy relationships between schools and families. Some of the most important
practices are detailed below.

First, family—centered practices are essential. These practices include a family
orientation rather than a child orientation, positiveness, sensitivity, and friendliness.
Second, schools should respect cultural diversity. Respect for diversity includes
understanding and appreciating cultural differences and considering them in trying to
forge relationships. A third essential practice is personal connections or informal
opportunities for interaction. Esler and colleagues cite examples like phone calls about
child successes, using first names, and personal contact between staff and parents.

The final major component of the Esler et al. (2002) model is actions. This
component is most well developed because into practice the attitude and relationship
components of the model. According to Esler and her colleagues, six components
comprise essential actions in forging home-school collaboration.

First, schools Should use the language of shared responsibility, which is inclusive,
solutions-oriented, and optimistic. Also, two-way communication between the home and
school is essential. However, Esler and colleagues acknowledge that this relationship has
historically been one-sided and that schools need to be proactive in creating and

maintaining channels of communication. Enhancing learning at home and school is the

third essential action and it has been studied extensively (e.g., Grolnick & Slowiaczek,
I994; Epstein, 1995). Fourth, schools should Share decision making at all levels, from
involving parents in interventions for individual students all the way up to school-wide
policy decisions. Fifth, collaborating with the community is important (Esler et al.,
2002). This action helps schools to access and mobilize community-based resources like
service groups and agencies. Finally, Esler, Godber, and Christenson note that a family-
school team is a prerequisite for enhancing relationships. These teams should include
community members as well as school professionals and families. Functions of these
teams include soliciting support from relevant parties, conducting needs assessments, and
monitoring programs and initiatives.

In summary, researchers have proposed various frameworks for understanding
home-school collaboration. Joyce Epstein’s 1995 model focuses on community
involvement and on defining the relationship in broad, inclusive terms. Her model has
been widely cited and celebrated because it expands the range of activities that contribute
to the home-school relationship and educational outcomes for children (Epstein, 1995).
In contrast, Diane Scott-Jones (l 995) presents a model of the home component of the
relationship between families and schools. She theorizes that parents engage in four
activities including valuing, monitoring, helping and doing schoolwork (Scott-Jones,
I995). Esler, Godber, and Christenson (2002) offer the most comprehensive model of
home-school collaboration focusing on attitudes, relationships, and actions. While this
model has the most conceptual strength, it still does not specify a psychological
mechanism by which the collaborative relationship affects families; scholars should strive

to fill this theoretical gap. In addition, parent perceptions of the attitudes, relationships,

l6

and actions that form the basis and the outcomes of the relationship have not been
examined and research should also address this deficit in the literature.
Review of Current Literature on Home-School Collaboration

The home-school relationship has been studied under a number of labels
including parent involvement and parent engagement. Four lines of research have
contributed most heavily to theory on the home-school relationship. One line of research
emerged in the literature on resilience. In this field, parent involvement is examined as a
protective factor that buffers children from negative outcomes and promotes well-being.
Second, research on parent involvement includes the study of child outcomes associated
with increased parent participation at school (e.g., Feuerstein, 2000). This data informed
policy makers and school professionals about why reaching out to families is important
(Esler et al., 2002). A third line of research that has shaped this field examines family
demographic factors that make parents likely or unlikely to be involved with schools. A
final area of research focuses on school or other organizational factors that increase
parent involvement and home-school collaboration.
Parent Involvement as a Protective Factor

Within a risk and resilience framework, parent involvement is examined as a
protective factor. In particular, Masten and Coatsworth (1998) note that having a close
relationship with a caring parent figure is one factor that promotes successful
development in children. Resilient children succeed academically and personally despite
the presence of significant risk factors in their lives (Mash & Dozois, 2003).

Resilience literature has expanded beyond children who succeed in the face of

adversity to exploring factors that promote success for all students. This success is

17

hypothesized to result from a combination of positive personal factors like being able to
seek out guidance and environmental supports like relationships with teachers and other
adults. Specifically related to parents, research shows that good quality parenting
prevents antisocial behavior among children exposed to high levels of adversity (Masten,
1994). In particular, Masten and her colleagues (I 999) followed a sample of children
from childhood into late adolescence and found that parenting contributed significantly to
resilience, which is marked by high adversity and adequate academic, behavioral, and
social competence. Specifically, they found that parenting predicted conduct in
childhood. Further, when Time One adjustment was accounted for, parenting in
childhood contributed significantly to social competence in late adolescence. In addition,
parenting in adolescence predicted competence in the academic, social, and conduct
domains in late adolescence.

Similarly, Pettit, Bates, and Dodge (1 997) found that supportive parenting acts as
a protective factor, predicting child adjustment over seven years. High quality parenting
indirectly moderated the effects of adversity on classroom adjustment for students with
behavior disorders and “mitigated the effects of family adversity on later behavior
problems” (p. 908). In particular, Pettit and colleagues found that supportive parenting
directly influences academic, behavioral. and social adjustment in kindergarten and sixth
grade and that it acts indirectly as a moderator of adversity. In the presence of socio-
economic status (SES) risk, sixth grade externalizing scores were higher when supportive
parenting was low. However, “When SES risk was absent, externalizing scores were

similar for low and high SP (supportive parenting) groups” (Pettit et al., 1997, p. 917).

18

Overall, research suggests that parent involvement in education serves as a protective
factor all students (Masten, I994; Masten & Coatsworth, I998).
Outcomes Associated with Home-School Collaboration

Within the field of home-school relationships, a second line of research examines
how parent involvement influences child academic outcomes including grades and
standardized test scores. Parent involvement is moderately associated with higher grades
(i.e., Fan & Chen, 2001; Keith et al., 1998). In a meta-analytical review of the literature,
Fan and Chen (2001) found a “small to moderate, and practically meaningful,
relationship between parent involvement and students’ academic achievement” (p. I). In
particular, they found that parent aspirations/expectations for the child’s academic
achievement moderates the relation between parent involvement and a global indicator of
academic achievement like grade point average.

In addition, Reynolds (1992) conducted a longitudinal study and found that parent
involvement in school is a significant predictor of future achievement. He examined how
parent, child, and teacher perceptions of involvement influence reading and math scores
on the Iowa Test of Basic Skills. Even after controlling for demographic information
including parent education, eligibility for a lunch subsidy, and grade retention, parent
involvement significantly contributed to academic achievement. Child reports of
involvement, rather than parent or teacher reports, accounted for the most variance in
standardized test scores.

Keith et al. (1998) also found a moderately large effect size in the relationship
between parent involvement and grades. Specifically, their model showed that a one

standard deviation change in parent involvement results in a .25 standard deviation

l9

change in tenth grade academic achievement. Results from Keith et al. also suggested
that gender does not affect the relationship between parent involvement and academic
achievement. However, they did find that the relationship differs by ethnicity; the
general model does not fit for Asian American students. Additionally, parent
involvement was seemingly more important for Native American students than the rest of
the sample and a standard deviation change in parent involvement was accompanied by a
.43 deviation change in grades (Keith et al., 1998).

In accordance with a broader view Of the successful outcomes, researchers have
also examined non-academic child outcomes associated with the home-school
relationship. Drawing conclusions about the link between parent involvement and
behavioral outcomes is challenging because the scales measuring student outcomes vary
widely, measuring everything from school behavior to dropout rates (e. g., McNeal,
1999). Despite the inconsistency in outcomes, overall, parent involvement seems to be
related to higher levels of school appropriate behavior (Grolnick & Slowiaczek, 1994;
Pettit et al., 1997).

For example, Grolnick and Slowiaczek (1994) found that for mothers, personal
involvement and cognitive/intellectual behavior are weakly but significantly correlated
with a child’s perceived academic competence and control understanding. In addition,
small correlations exist between mother-reported activities supporting “the dedication of
psychological resources in the context of positive affect” and self-regulation (Grolnick &
Slowiaczek, 1994, p. 241). In addition, mother and father cognitive/intellectual

behaviors, which include cultural activities at home like reading the newspaper and

20

talking about current events, are weakly correlated with a child’s perceived competence
in the academic area (Grolnick & Slowiaczek, 1994).

Using a social capital framework, McNeal (1999) also examined the impact of
parent involvement on non-academic outcomes. His research suggested, “parent
involvement in general significantly reduces incidents of problematic behavior” (McNeaI,
1999, p. 129). McNeal also teased apart the results for students from different racial
backgrounds. His research indicated that parent involvement as social capital
“consistently affects behavioral outcomes” (like truancy and dropping out) for Caucasian
students, has less impact on African-American students and almost no impact on
Hispanic and Asian students (p. 131).

Family Factors Affecting Parental Involvement

A third area within the literature on home-school relationships is family factors
that make parents likely or unlikely to be involved at school. For instance, family
configuration predicts parent involvement at school; single parents are less involved in
school-based behaviors like volunteering in the classroom and PTO attendance than two-
parent families (Kohl, Lengua, & McMahon, 2000). Parental education also predicts
involvement. In particular, research suggests that more educated parents are more likely
to have high levels of involvement (Kohl et al., 2000; Shumow & Miller, 2001). Third,
race predicts involvement. Caucasian and African American parents are more likely than
Asian American and Hispanic parents to engage in traditional parental involvement
activities like volunteering at school. Desimone (1999) hypothesizes that language
barriers may prevent involvement at school. Race also predicts communication about

school. Caucasian students report the highest level of communication about school and

2]

Asian American students report the most talk about educational aspirations (Keith et al.,
1998)
School Factors and Initiatives

A final strand of the research on home-school relationships examines school
factors and initiatives that increase parent involvement at school. Most Of this research
offers school staff specific school-wide strategies for involving parents in the classroom
and school (i.e., Baker, 2000). This literature emphasizes the importance of including
families on a school-wide level through practices like fostering parent’s role as decision
makers (e.g., Hara & Burke, 1998).

Hoover-Dempsey, Bassler, & Brissie (I987) offer one example from this area of
research. They used multiple regression to examine the factors that contribute to parent
involvement and found that the most consistent factors in predicting parent outcomes
were teacher efficacy and school socioeconomic status (Hoover-Dempsey et al., 1987).
Instructional coordination between grades, principal perceptions of teacher efficacy, and
parent tutoring at home also contributed significantly to parent involvement outcomes.
This research suggests that school-level variables can play an important role in
influencing the families of their students.

Paradoxes and Gaps

Although the field of home-school relationships is burgeoning, a number of
paradoxes and gaps still exist within the literature. First, while home-school
collaboration is intuitively appealing and a mainstay in educational policy, the
phenomena is not well understood. For instance, the Elementary and Secondary

Education Act specifies that school districts must earmark at least one percent of their

money for promoting parent involvement (Baker, 2000). Esler and colleagues (2002)
also list many other federal education bills that devote funds to increasing parent
involvement. Despite this recent interest and investment in parent involvement, the
relationship between families and schools remains unclear (Christenson & Godber,
2001)

One factor that contributes to this lack of understanding is the plethora of
atheoretical research devoted to helping schools facilitate the home-school relationship.
Despite the lack of empirical support, respected researchers within school psychology
and education are quick to present strategies to foster home-school collaboration. For
example, Esler, Godber, and Christenson (2002) moved from a basic review of the
literature directly into presenting strategies to improve the home-school relationship.
While their strategies are congruent with best practices in education and psychology, they
are not based on a solid psychological model. This could be related to the applied nature
of the field of school psychology and the relative newness of this construct.

Another factor that contributes to this lack of clarity, and a second paradox within
the field, is that despite years of study, researchers have yet to agree upon a clear
mechanism by which home-school collaboration operates. Various authors within the
field of school psychology have proposed plausible explanations for the relationship, yet
none of these models has become the dominant paradigm.

Perhaps the most prominent theory is Epstein’s (1995) model of
home/school/community partnerships, which is based on the idea that families, schools,
and communities exist in relationship to one another. Epstein presents six components of

the relationship including parenting, communicating between the home and school,

to
'JJ

family members volunteering at school, fostering learning at home, Shared responsibility
for decision making in the school, and collaborating with the community. Despite the
fact that her model is widely cited, researchers continue to propose and examine other
models (e.g., Swap, 1990).

Related to the second paradox about the lack of a theoretical model is the lack of
empiricism in the field. While researchers have proposed models (e.g., Esler et al.,
2002), rarely are they statistically validated. Grolnick and Slowiaczek (1994) present an
exception to this trend in offering a three—component model of how parent involvement
influences learning. Specifically, they examined a multidimensional conceptualization of
parent involvement and hypothesized a model in which children’s motivational resources
(including perceived competence, control understanding, and self—regulation) mediated
the relationship between parent involvement and academic performance. Three hundred
and two eleven- through fourteen-year-olds and their parents and teachers in a
predominantly Caucasian area participated in this study. Factor analysis validated their
three-component model of parent involvement. The three components include: behavior,
which includes involvement activities like volunteering at the school, personal
involvement, or the parent valuing the school and providing the student with resources,
and cognitive/intellectual, which involves the parent providing the child with cognitively
stimulating activities.

In addition to presenting a three-component model, Grolnick and Slowiaczek’s
(1994) results suggested that children’s motivational resources acted as a mediator in the
relation between parent involvement and school performance. Specifically, maternal

behavior and cognitive/intellectual support uniquely predicted perceived competence and

24

control understanding, which in turn predicted school performance. Based on the
findings, the authors recommend that researchers use multi—dimensional indices of parent
involvement in education and that scholars examine how parent involvement influences a
variety of outcome measures. Despite this well-constructed and empirically sound
research, Grolnick and Slowiaczek’s model has not risen to the forefront of the field.

A fourth paradox within this field is that the lack of conceptual clarity has in no
way hindered research on outcomes of the home-school relationship. As noted above,
this literature demonstrates a moderate association between parent involvement and
student grades (i.e., Fan & Chen, 2001; Keith et al., 1998). In addition, research on
home—school collaboration also includes behavioral outcomes (Pettit et al., 1997;
Grolnick & Slowiaczek, 1994). Again, this research has proceeded despite a lack of
conceptual and empirical clarity.

Addressing the Gaps

The previous section highlights a number of incongruities in the field. The
present study is designed to address some of these gaps within the literature. As noted in
the previous section, much work remains to remediate these deficits.

First, a conceptually sound model of the home-school relationship must be
empirically validated to link theory and research. As noted before, one model that
deserves further study is Esler and her colleagues’ (2002) model ofhome-school
collaboration. Second, the field of home—school relationships needs to devote time and

energy to understanding the underlying psychological mechanism and to specifying a

psychological theory by which these relationships Operate. Third, this field needs to

25

widen its lens beyond child outcomes to parent outcomes associated with the relationship.
Finally, moderators of the newly specified home-school relationship should be examined.
Model of the Home-School Relationship

As noted earlier, Esler and her colleagues (2002) offer a comprehensive,
appropriately focused, and testable theory of home-school relationship involving
attitudes, relationships, and actions. This model of home-school collaboration has three
main advantages including flexibility, dynamism, and an affective emphasis. Despite the
many strengths of the model, two serious weaknesses exist. This model does not account
for the psychological mechanism by which home-school collaboration influences families
and it under-represents parent perceptions of the relationship.

As noted above, one advantage of the model is that the authors flexibly define
involvement so that it does not require resources like time during the day and post-
secondary education. Specifically, many models of the home—school relationship call for
parents to participate at the school, serving on the PTA and volunteering in the classroom
(e.g., Feuerstein, 2000). While these activities are obviously beneficial to students and
help to establish relationships (e.g., Marcon, 1999), they are fraught with barriers for
some families. For instance, in many homes, all of the adults work full time and cannot
take time off during the day. However, research by Bempechat (1992) suggests that
possessing fewer resources does not necessarily predict less educational support by
parents (see also Scott-Jones, 1995).

The second strength is that the Esler et a1. (2002) model is dynamic and
responsive. Changes in any component can lead to improved connections between

families and schools. For instance, the model suggests changes can result from actions

26

and attitudes as in the case of parents becoming involved in a parent-school team and
changing their perceptions of their role in educating their children. Also, the model
captures the fluid nature of the relationship. Rather than a hierarchical conceptualization
with schools providing opportunities for at-school involvement and parents choosing
whether or not to respond, this model emphasizes relationship building and attitudes as
well as school- and parent-initiated actions.

Third, this model has a strong emphasis on the affective component of family-
school relationships, which goes hand in hand with the reduced emphasis on parent
actions requiring resources like time, money, and post-secondary education. Esler and
colleagues (2002) note that valuing education coupled with communication, engagement
in decision-making, and enhancing learning at home are vitally important and sufficient
to benefit students. In stressing the affective nature of the model, Esler et al. also remove
pressure and blame from parents who have less education or familiarity with formal
education. Their model suggests that believing in the value of education, conveying that
belief to children, and staying involved in the lives of children can make all the
difference.

One major weakness of the Esler et al. (2002) model is that it does not specify a
psychological mechanism by which the home-school relationship influences outcomes.
The authors describe the necessary components and conditions, but do not present the
specific internal processes that create change. Future work in this area should strive to
explain how home-school collaboration influences families.

A second weakness resides in the attitudes component, which neglects parent

attitudes towards the school. Parent perceptions and appraisals of their relationship with

27

the school are not presented in this Best Practices in School Psychology chapter on
supporting home-school collaboration (Esler et al., 2002). While the chapter includes
teacher perceptions of sharing responsibility for the developmental and learning
outcomes of children and a framework for school psychologists to assess and develop
constructive attitudes, it does not include parents beyond their ambiguous role as non-
specific stakeholders. Further, the authors describe empowering parents and teachers to
change their attitudes, yet they do not explicitly describe how to measure let alone change
parent perceptions of the relationship. Future research should examine parent perceptions
of actions, relationships, and attitudes that comprise the home-school relationship.
Social Support

As previously noted, one of the major challenges within the study of home-school
interactions is the fact that a mechanism explaining the relationship has not been
specified. Psychology Offers a potential mechanism with the construct of social support.
Social Support: An Explanatory Mechanism for the Home-School Relationship

Although social support is a seemingly straightforward construct, a vast array of
definitions exists within the literature. Defining social support in such a way as to
include the myriad forms of support (e.g., one’s social network, professionally-affiliated
colleagues, and professional service providers) while still maintaining a coherent
definition has proven to be very challenging.

For the purposes of this text, Cauce, Felner, and Primavera’s 1982 summarization
of Caplan’s (1974) definition of social support will be used. Cauce and colleagues
(1982) define social support as “the range of significant interpersonal relationships that

have an impact on an individual’s functioning” (p. 418). This umbrella definition

28

highlights essential human relational needs. This definition also suggests that social
support plays an active role in functioning. However, one area where it falls short is in
limiting relationships to interpersonal relationships between one or more individuals.
This definition excludes support provided by organizations as a whole, such as churches,
employers and schools (Demaray & Malecki, 2002; House, 1981). For this study, social
support is hypothesized to flow from schools and school staff members to families and
parents.
History of Social Support

Conceptual and methodological complexity within the field stems from the fact
that this research arose from three very different bodies of knowledge (Sarason, Sarason,
& Pierce, 1990). Areas of research that gave rise to social support include the stress and
coping literature from within the medical field, research on sociology and community
psychology, and the attachment literature. Briefly, stress and coping literature examined
people who maintained mental health despite the presence of significant stressors
(Folkman & Lazarus, 1980). Sociology and community psychology examined
individuals’ relationships with others and described the benefits of social connectedness.
Finally, social support arose from John Bowlby’s work on parenting and attachment.

Research on social support first emerged within the field of clinical medicine
(Lazarus, 1980; Sarason et al., 1990). Within that literature, John Lazarus examined the
impact of stress on veterans from World War II and the Korean War and noted that stress
significantly impairs functioning in some contexts (Lazarus, 1966).

Folkman and Lazarus (1980) extended research on social support to community

settings. In their study of middle-aged persons, Folkman and Lazarus found that both

29

problem- and emotion-focused coping are essential in stressful episodes. Problem-
focused coping refers to “cognitive problem solving efforts and behavioral strategies for
altering or managing the source of the problem” while emotion-focused coping refers to
“cognitive and behavioral efforts directed at reducing or managing emotional distress” (p.
224-225). They also examined the characteristics of stressful situations and found that
context and an individual’s appraisal of the situation are the most important factors in
coping.

Cassel (1976) too Offered a medical perspective on social support. He used an
epidemiological lens on stress and coping, emphasizing social relationships as a factor
that increases protection against environmental disease agents (Cassel, 1976).
Specifically, he noted that psychosocial processes could act as stressors or protective
factors. In the same text, Cassel urged the medical community to turn its attention to
individuals at risk due to poor fit within their social milieus.

Cobb (1976) also studied social support in a medical context, conceptualizing it as
a factor that protected patients against negative outcomes in times of illness. He
examined literature in which social support mediated the relationship between a disorder
and a variety of outcomes rather than examining social support as a main effect (Cobb,
1976; Sarason et al., 1990). Cobb summarized research on a variety of conditions
ranging from low birth weight and arthritis to depression and alcoholism. He noted a
strong and persistent mediation effect with social support reducing the amount of
medication required, accelerating recovery, and increasing compliance with prescribed

regimens.

30

Sociology and community psychology also gave birth to the construct of social
support (e.g., Auerbach & Kilmann, 1977). These related fields describe social support
in terms of social connectedness and social webs. For instance, Auerbach and Kilmann
reviewed how support relationships influenced psychiatric crisis interventions. They
suggested that the field adopt a pre-surgery model with group and individual sessions of
relaxation and support for these patients in crisis. Auerbach and Kilmann also noted that
successful crisis intervention plans already include a coping or social component.

Whitcher and Fisher (1979) also studied the impact of nurse—initiated therapeutic
or interpersonal touch in hospitals. In a controlled experiment, Whitcher and Fisher’s
research suggested that female patients experience positive “affective, behavioral, and
physiological” reactions to the touch (p. 87). In particular, female patients reported less
anxiety, were satisfied with pre-operative instruction, liked their nurses, read the
preoperative literature, and reciprocatcd touch. Their vital statistics were also closer to
normal than those of patients who did not experience touch. On the other hand, males
demonstrated negative reactions to therapeutic touch. Based on their research, Whitcher
and Fisher recommend that nurses make a concerted effort to use therapeutic touch with
female patients. Overall, they suggest that support dramatically improves outcomes for
female patients.

Social support also arose from attachment theory derived from Bowlby’s (1969;
1980) work on the relations between infants and their mothers. His 1969 book
introduced the biological and emotional components of social support. Initially, Bowlby
viewed attachment from a psychoanalytical lens, describing the relationship between the

mother and child in terms of needs and drives. Bowlby also presented a biological case

31

for attachment based on his studies of animal behavior. In his later iteration of
attachment theory, Bowlby presented attachment as a normal, biologically-based, and
developmental]y-appropriate process based in cognitive psychology (1980). In his 1980
text, Attachment and Loss, Bowlby presents caregiving as “protecting the attached
individual” (p. 40). He elaborates, stating that caregiving is behavior shown by an adult
towards a child, or behavior “shown by one adult towards another, especially in times of
ill health, stress, or Old age” (p. 41). Although he does not label this caregiving behavior
social support, the idea of one individual supporting another in times of stress is
remarkably consistent with social support.
Theoretical and Conceptual Challerlges Associated with Social Support

Various researchers have concluded that theoretical and historical openness
creates conceptual and methodological challenges (Tardy, 1985; see also Cutrona &
Russell, 1990, Sarason, et al., 1990; Thoits, 1982a). As suggested by the definition and
history, the first problem plaguing the field is the lack of a clear definition or framework.
lntuitively, the term social support suggests the sense of well-being people derive fiom
their relationships with others. At its most basic level, social support does fit this
definition. In particular, social support as emotional support is the well-being people
derive from perceptions of others’ regard for them (House, 1981; Thoits, 1982a).
However, beyond this basic definition, social support becomes far more nebulous and
fraught with conflict (Thoits, I982a).
Theories of Social Support

As suggested by its diverse origins, social support has been conceptualized in

many ways throughout its history. In order to bring conceptual clarity to the field and to

32

summarize a diverse literature, Tardy (1985) presented a five-component model of social
support integrating the work of many scholars. His model includes direction, disposition,
description and evaluation, content, and source.

As noted above, the first component hypothesized by Tardy (1985) is direction.
Direction refers to whether an individual is giving or receiving social support. Direction
also encompasses the idea that the relationship can be reciprocal or unidirectional. For
instance, a counselor or clergy can be a source of support without receiving support from
the relationship.

Disposition is the second component of social support; it includes both the
availability and the utilization of support (Tardy, 1985). Specifically, availability refers
to the quantity and quality of available support. Utilization refers to persons availing
themselves of the available support.

Third, Tardy (I985) summarizes the description and evaluation of social support.
Description refers to models of social support that describe existing support resources.
Evaluation is a more abstract component and refers to individual’s perceptions of
support. It can also refer to satisfaction with levels of support (Demaray & Malecki,
2002)

The fourth component of social support is content (Tardy, 1985). Significant
diversity exists within conceptualizations of the content of social support (e.g., Demaray
& Malecki, 2002; Thoits, 19823). For instance, Thoits (1982b) conceptualizes social
support from an emotional perspective. Her measure taps into belonging, friendships,
and membership in clubs or religious affiliation. House (1981) broadens the notion of

social support, viewing it as extending into actions and information as well as

interpersonal relationships. The four components specified by House (198 I) and
presented by Tardy (1985) are emotional support, appraisal support, instrumental support
and informational support. An extended presentation of these components follows
below.

Fifth, research on social support includes its sources (Tardy, I985). Tardy
highlights family, close friends, neighbors, co-workers, community, and professionals as
providers of support. Some researchers examine sources in terms of their presence in the
lives Of others, while other researchers are concerned with the characteristics Of those
who comprise the networks. This conceptualization leaves out the characteristics or
qualities of organizations, such as schools, that provide support. Different sources of
support provide different components based on their roles in the life of the person being
supported (Cauce & Srebnik, 1990). Specifically, Cauce and Srebnik note that sources of
support vary in terms of their intimacy and their relevance to daily life. For instance,
schools often have little intimacy with families but high relevance for daily life. On the
other hand, fiiends have both high intimacy and high levels of relevance to daily life.
Model of Social Support

In this study, social support will be conceptualized in terms of House’s 1981
book, Work Stress and Social Support. Tardy’s 1985 model subsumes the content Of
House’s model, but Tardy advocates for assessing more dimensions of the construct. In
order to focus much needed attention on the content of the social support provided by the
schools to parents, this study largely disregards direction, disposition, description and

evaluation, and sources of support.

According to House, emotional support is the most important component of social
support. He notes that emotional support pervades intuitive and empirically supported
definitions of social support. Also, House states that emotional support is the only
wholly positive component whereas the other components can have negative
implications. For instance, instrumental support can lead to excessive dependence (Cobb,
1976). While emotional support is most central to the construct, instrumental, appraisal,
and informational support are also essential (House, 1981).

House (1981) defines instrumental support as “behaviors that directly help the
person in need.” Instrumental support is most clearly differentiated from emotional
support by its emphasis on actions rather than affect or cognition. Examples of
instrumental support include giving others time or goods. Specifically, providing parents
with transportation to school conferences or a school providing a child with school
supplies are examples of instrumental support.

Informational support is the provision of information that helps others “in coping
with personal and environmental problems” (House, 1981, p. 25). House differentiates
informational support from instrumental support by stating that informational support
allows parents to help themselves (1981). Examples of informational support include
notifying parents about low cost summer camps or providing them with job skills
training. Another example is a school informing parents about how to foster early
literacy skills in their young child.

The final component of social support is appraisal support. Appraisal support
includes information about an individual or an individual’s performance (House, 1981).

Specifically, it provides persons with information for self-evaluation or social

comparison. One example is supervisors providing employees with feedback about their
job performance (House, 1981). In a school system, this support could include schools
elucidating their expectations of parents and telling them how important they are for
education.

Although the four types of support are clearly separate, natural divisions exist
between emotional and appraisal and information and instrumental support (Thoits,
1982a). Viewing social support in this way simplifies the content and produces a
potentially more parsimonious model. Further, previous scholars have noted this
dichotomy. First, Thoits (1982a) argued convincingly that social support can be
aggregated into socioemotional aid and instrumental aid. Also, House (1981) noted that
emotional and instrumental support have the most weight as separate constructs. Thus, in
this study, social support is theorized to include emotional support (which includes
appraisal support) and instrumental support (which includes informational support).
Mechanisms of Social Support

Scholars posit that social support acts in two ways to support the health and
wellbeing of individuals (House, 1981). In particular, social support is thought to act as a
main effect and to buffer the relation between stress and adjustment.

Social Support as a Main Efl'ect

Social support directly influences levels of stress or adjustment (House, 1981).
Specifically, it can directly decrease the amount of stress and social support can increase
perceptions or measured levels of adjustment. In addition to influencing the affective
component of the situation, social support can also act in the coping process, managing

the problem that is causing the distress (Cutrona & Russell, 1990).

36

To illustrate social support as a main effect, consider the example of a family
experiencing financial pressure due to job loss. Social support can lower stress related to
financial insecurity. It can also increase adaptive feelings like hope or optimism about
the future. Finally, social support can help someone secure a new job. In an ideal world,
social support would serve all three functions.

Many groups of researchers have examined social support as a main effect (e.g.,
LaRocco, House, & French, 1980 and Wenz-Gross, Siperstein, Untch, & Widaman,
1997). For example, Wenz-Gross, Siperstein, Untch, and Widaman (1997) examined
how support from family, other adults, and peers influences child outcomes. They found
that family support contributes directly to academic self-concept and liking of school
(Wenz-Gross et al., 1997). Peer support has more varied results. Specifically, problem-
solving support decreases social self-concept while companionship significantly increases
social self-concept (Wenz-Gross et al., 1997). These findings illustrate the direct role
social support plays in the adjustment of students. These results also suggest that domain
is important, and that support is more likely to play a role when the source and type of
support are closely aligned with the outcome measure.

Dubow, Tisak, Causey, Hryshko, and Reid (1991) offer another example of the
direct effects of social support on adjustment. They conducted a three-year study of third
through fifth graders. Using multiple regression, they Showed that initial social support
influenced academic achievement and that initial family support predicted teacher-rated
child competency (Dubow et al., 1991). Dubow and colleagues also ascertained that

increases in social support predicted improvement in behavior problems, teacher-rated

37

competency, and grades. Overall, this study suggests that support from parents, teachers,
and peers significantly affects adjustment.

Recent conceptualizations of social support have examined it as a protective
factor for all students, regardless of the presence or absence of stress (i.e., Demaray &
Malecki, 2002). This fits with the current positive psychology and resilience movement,
which examines factors that promote health and wellbeing in all people. For instance,
Demaray and Malecki (2002) studied third through twelfth graders and found
relationships between perceived social support and positive outcomes including social
skills, self-concept, and adaptive skills. In particular, Significantly different results
emerged for students with low, medium, and high levels of support (Demaray & Malecki,
2002)

Social Support as a Buffer

A vast majority of research examines social support as a buffer or miti gator of
distress (e.g., Cobb, 1976, House, 1981, Thoits, 1982a). This theory emerged from
research suggesting that certain people were less affected by numerous life changes than
others with similar patterns of stressors (Thoits, I982a). Buffering relationships are
hypothesized to occur when the relationship between an independent and dependent
variables differs in the presence of a third variable. In particular, “a moderator interacts
with a predictor variable in such a way as to have an impact on the level of a dependent
variable” (Holmbeck, 1997, p. 599). This type of relationship is also known as
moderation.

Most often, the relationship between a stressor and an adjustment outcome is

examined with social support as the moderator. For example, House (1981) stated that

38

the harmful impact of stress on health is reduced as social support increases. He
hypothesized that support has the greatest effect for people under stress and little to no
effect in the absence of stress. As Cohen and Wills (I 985) suggested, an interaction
between stress and social support is hypothesized such that the high levels of social
support moderate the negative effects of stress on adjustment.

Many studies examine social support as a buffer (e.g., Cobb, 1976; LaRocco,
House, & French, 1980; Thoits, 1982a). Cobb initially posited this model. He examined
social support in the medical literature, noting that social support acts as a protective
factor against a variety of pathological states ranging from low birth weight to psychiatric
illness.’ Cobb intentionally omitted studies examining social support as a main effect in
his review of the state of the field.

LaRocco, House, and French (1980) examined social support as a buffer in the
relationship between occupational stress and health. In a study of 636 men, they found
that the relationship between the perceived job stresses (e. g., role conflict and future
ambiguity) and the outcomes of depression, irritation and somatic complaints was
strongly buffered by social support. Only in the outcome measure of anxiety did social
support not act as a buffer. In addition, the relationship between the job-related strains of
job dissatisfaction, boredom, and workload dissatisfaction and the health variables also
revealed the buffering effects of social support.

Social Summit Provided by the Schools Influences F amilv — School Relationships

As previously noted, the relationship between families and schools remains

”unclear within research on education and school psychology (i.e., Grolnick &

Slowiaczek, 1994; Esler et al., 2002). Psychological theory, particularly the rich theory

39

sustaining the construct of social support, has the potential to elucidate this murky
interaction. In particular, literature suggests that social support might be one mechanism
explaining the relationship between homes and schools, and second, that this relationship
has the potential to influence parent outcomes.

Three ideas from the social support literature suggest that school-provided support
might influence family-school relationships. Specifically, LaRocco, House and French
(1980) posit that specific types of Stress or strain (e.g., parent stress with school) are
likely to be affected by a limited set of closely related sources of support. Second, House
(1981) opines that informal and non-professional forms of social support are important
and perhaps even more important than formal sources of support. Third, Vaux (1988)
found that social support provided by the schools influences child outcomes. Given this
link, it is possible that social support also has the potential to influence family outcomes.

First, LaRocco, House and French (1980) offer a specificity of domain theory in
social support. They hypothesize that specific types of stress are more likely to be
reduced by a limited set of closely related resources. LaRocco, House and French extend
this argument to suggest that both proximal stressors and proximal sources of support
affect outcomes. Taken further, this could suggest parent stress about school would be
reduced by social support that ameliorates school-related stressors.

LaRocco, House, and French’s specificity of domain theory is indirectly
supported by research and theory from Dubow, Tisak, Causey, Hryshko, and Reid
(1991). Dubow and colleagues (1991) assert that schools and school personnel offer up
specific types of support related to school. In particular, they note, “Teachers may be

more often viewed as providers of informational and tangible support specific to school-

40

related concerns”(p. 585). Taken to its logical conclusion, this research suggests that
support in the school domain will impact school-related outcomes.

The specificity of domain theory is also supported by Cutrona and Russell’s
(1990) theory of optimal matching. They hypothesize that desirability, controllability,
duration of consequences, and life domain necessitate certain kinds of social support.
Concerning desirability, Cutrona and Russell note that desirable events like a job change
are associated with anxiety while undesirable events produce depression; further, more
social support is necessary to cope with undesirable events. Second, controllable events,
like moving, need instrumental support while uncontrollable events like a divorce or the
transition to high school require emotional support. Third, duration influences matching;
longer-term events require more long-term and tangible support. Finally, the domain
affects what type of support is most necessary. Although Cutrona and Russell do not
explicitly address schools, their theory can be extended to this area in suggesting that
school-related support most influences school-related outcomes.

In summary, optimal matching theory suggests that certain types of social support
are more important in the presence of certain specific stressors and that social support is
most effective when matched by domain or by individual need. Extending this theory to
the schools, I propose that the social support provided by the schools will influence
parents’ relationship with the schools.

House (1981) offers a second avenue of connection between family-school
relationships and social support. In Work Stress and Social Support, he suggests that
informal and non-professional forms of social support are important and perhaps even

more important than formal sources of support. House offers three arguments to support

41

this assertion. First, House notes that when surveyed, people spontaneously list informal
sources of support like family and fiiends over more formal sources like physicians,
clergy or psychologists.

A second rationale for studying informal sources of support is that they are most
consistent with a prevention focus (House, 1981). If fiiends, families, and schools offer
adequate amounts of support, then outside providers like welfare services or social
workers may not be necessary. House notes that relatively small expenditures of
resources can pay off enormous dividends in terms of healthier group processes and by
extension, families. In a school setting, improved social support could prevent parents
from needing the support of community mental health or similar providers.

Third, House (1981) reports that informal sources of support can be very effective
in reducing stress, improving health, and buffering the impact of stress on health. He
hypothesizes that informal sources of support may be effective for two reasons. First,
informal support may come with more equality, which makes it more palatable to the
recipient. Second, informal support tends to be spontaneous rather than planned, which
makes it seem more genuine and thus potentially more effective.

Overall, House (1981) argues that social support from colleagues and supervisors
at work reduces stress, prevents health problems and helps people to adapt to stresses at
work. Since schools are in a position to fulfill some of the same roles, they should be
examined as a potential source of social support for parents. In addition, House’s theory
suggests that perhaps schools should strive to foster equality and spontaneity in the

support they provide.

42

A third reason social support provided by the schools might explain the home-
school relationship and influence parent outcomes is the socialization role schools play in
the lives of children (Demaray & Malecki, 2002; Scott-Jones, 1995). Specifically, school
policies and practices influence the social lives of children and it follows that the families
of Students might also be affected by their interaction with the school (Vaux, 1988).

As Vaux (1988) notes, “School policies often influence the social relationships of
youth (p. 267).” Specifically, Vaux addresses how school policies influence child
outcomes like independence and social integration. He notes that schools also “shape
social networks and opportunities for social support (p. 268).” While Vaux argues
convincingly that schools influence the lives of children, he neglects to mention how
school practices and policies could affect parents. He examines the direct effect of
school on children and work on adults, but neglects the more proximal component of how
contact with the schools influences families. This is problematic because parent
outcomes are more proximal to the home-school relationship than the more distal
outcome variable of child adjustment (Dubow, Tisak, Causey, Hryshko, & Reid, 1991).
Broadening the Lens to Include Family Outcomes

As previously noted, one of the paradoxes within the field is that research on the
home-school relationship has focused almost exclusively on child outcomes despite the
fact that children comprise only one part of the equation. The home-school relationship
also affects parents. In fact, parents are more central to the relationship than any other
party, which suggests that parent perceptions of the relationship and parent-related

outcomes should be examined. Although a few scholars have examined parent

43

perceptions of the school and parent outcomes (e.g., Goldring & Shapira, 1993; Griffith,
1997), these studies do not paint a complete picture of the home—school relationship.

Existing literature begins to explain outcomes of the home-school relationship but
clearly more work remains. Most current research on parent outcomes related to school
is housed under parents’ satisfaction with schools resulting from school choice (e. g.,
Goldring & Shapira, 1993; Hausman & Goldring, 2000). In one study of school choice in
Israel, Goldring and Shapira (1993) examined parent satisfaction with schools of choice
in terms of eight areas including academic, social, and educational philosophy. In order,
the factors that are most highly associated with parent satisfaction are parent involvement
with the school, compatibility with expectations, parent empowerment, and parent
education level. Their results also suggest that socioeconomic status is an important
factor in this relationship.

James Griffith (e.g., 1997; 1998; 2000) also examines parent outcomes pertaining
to their children’s schools. Specifically, he used regression analyses to examine the
predictive power of student and parent perceptions of the school (including things like
instruction and climate) on student and parent outcomes including involvement and
satisfaction (Griffith, 2000). In his study of one hundred and twenty-two elementary
schools, his results indicated, “consensus among student and parent perceptions regarding
the school environment was significantly and positively correlated with their evaluation
of the school environment” (2000, p. 35). Based on the findings, Griffith recommends
that researchers evaluate consensus between students and parents in order to influence
parent outcomes. He also suggests that education researchers turn to the social

psychology literature for support in establishing a supportive and warm school climate.

44

Testing a Conceptual Model of the Home-School Relationship

As noted earlier, Esler et al. (2002) created a model of home-school relationship
with attitudes, relationships, and actions. While the actions component, or parent
involvement as traditionally defined, has been well studied and is well-understood,
family attitudinal and relational outcomes have been largely ignored. This gap poses a
problem because the Esler model and other literature on home-school relationships
indicate that attitudinal and relational outcomes are as important as actions.

Attitudinal and relational outcomes, while largely neglected, are crucial
components of the home-school relationship and will be elucidated further below. In
particular, the attitudes component will be represented by parent satisfaction with school,
parents’ appraisal of their child’s school and their relationship with the school. The
relationship component of the model will be assessed through parent trust of the school.
Actions will be defined according to traditional definitions of parent involvement and will
thus be measured as parent activities that support learning.

Family Satisfaction with School: Attitudes

Family satisfaction with school is a family’s affective appraisal of their child’s
school life and experiences. It includes relationships with teachers and other staff
members as well as a family’s sense of the quality of their child’s educational
experiences (Griffith, 2000).

Family satisfaction with schools has been conceptualized and measured in a
number of ways. For instance, Hausman and Goldring (2000) ask parents to assign their
schools a letter grade ranging from A for highly satisfied to F for highly dissatisfied.

James Griffith (2000) conceptualizes parent satisfaction with school in terms of four

45

components including the child’s attitude towards learning, the quality of education, and
“whether the parent would recommend the school to others” (p. 57). Goldring and
Shapira (1993) offer a more wide—ranging definition of parent satisfaction centered
around satisfaction in particular domains including academic and the general school
atmosphere.

Little research examining family satisfaction with school exists and researchers
tend to create and use their own measures of the construct. For instance, Goldring and
Shapira (1993) examined the impact of education level, trust, congruence with
expectations, and involvement on parent satisfaction with schools in Israel. Parent
satisfaction was assessed using a new, eight-item, 4-point Likert-type scale that included
academic, social, citizenship, general school atmosphere, values, educational philosophy,
developing individual potential, and the curriculum. Three hundred and thirty-seven
parents from four schools in Israel participated in the study. Results showed that together
education, trust, congruence with expectations, and involvement accounted for 20 percent
of the variance in parent satisfaction. Specifically, education level was negatively
associated with satisfaction, while the other three variables are positively associated with
satisfaction. Further, the researchers separated the sample by education and found that
for parents with less education, involvement is the only variable that significantly
predicts satisfaction. For more highly educated parents, compatibility, empowerment,
and involvement were significantly associated with satisfaction. Based on these findings,
Goldring and Shapira suggest that compatibility with their expectations is the most
important factor in determining parent satisfaction, followed by empowerment and

involvement. This relationship varies by SES with more wealthy parents experiencing

46

less satisfaction. This study aside, the construct of parent satisfaction with school has not
been widely examined within the literature.
Parent Trust: Relationships

Parent trust is a new construct within the home-school collaboration literature. It
has been examined as a process that facilitates collaboration (e.g., Godber & Christenson,
1999), as an outcome of the school climate (e.g., Soodak & Erwin, 2000), but not as an
outcome of the relationship between families and schools. For this study, parent trust
will be defined according to Adams and Christenson’s 2000 study as “confidence that
another person will act in a way to benefit or sustain the relationship, or the implicit or
explicit goals of the relationship, to achieve positive outcomes for students” (p. 480).
Adams and Christenson cite Rempel, Holmes, & Zanna’s (1985) study in defining trust
as a progression from predictability to dependability to faith between the two parties.

Smith, Atkins, and Connell (2003) examined how parent and child trust of
teachers related to academic outcomes for African American children. Smith and
colleagues defined parent trust solely in terms of racial-ethnic attitudes. Specifically,
their two trust items for parents were summarized as “Trust friends of other races” and
“Trust other races” (p. 166). This survey of 98 parents of fourth graders revealed that
parent attitudes significantly affect child racial-ethnic attitudes and achievement. Path
analysis results indicated that parent education was significantly related to child
achievement and that racial-ethnic pride was related though not significantly. Overall,
this study suggested that parent, child, and teacher trust influences child outcomes and

deserves further study.

47

The best example of research on parent trust comes from Adams and Christenson
(2000). They examine parent trust as a condition that contributes to collaboration. In
their study, they described trust in elementary, middle, and high school students,
examined factors that contribute to trust, studied whether amount of trust is mediated by
the amount of contact between families and schools or by demographic information, and
examined whether parental trust is related to child academic outcomes.

Adams and Christenson used surveys from 1,234 parents and 209 teachers in a
predominantly Caucasian area to examine this phenomenon. Their data indicated that
parent trust of the school is highest in elementary school and that parent trust of teachers
is significantly higher than teacher trust of parents in both elementary and high school.
Further, using an open-ended question, Adams and Christenson (2000) found that
improving communication was recommended by both parents and teachers to improve
levels of trust. The authors also noted that perceived quality of the relationship was a
better predictor of trust than demographic information or level of contact. Finally, parent
trust was significantly related to high school credits earned per year, GPA, and
attendance, but not to standardized achievement test scores for younger students.

Parent Involvement: Actions

Since attitudes and relationships are explicitly addressed by the other two
outcome measures, the actions component will be defined somewhat narrowly as
traditional, parent-initiated involvement activities (Kohl et al., 2000). Grolnick and
Slowiaczek (1994) define parent involvement as “the dedication of resources by the

parent to the child within a given domain” (p. 238). This broad definition encompasses

48

parent personal, cognitive/intellectual actions and behaviors. This portion of the text will
focus only on the behaviors component of their measure.

Grolnick and Slowiaczek (1994) define parent actions in terms of observable
behaviors. To assess parent behaviors, they created a scale assessing school-related
actions including attending parent-teacher conferences, open houses, and school
activities. Their measure also asked if parents had met teachers. Grolnick and
Slowiaczek hypothesized that parent behavior directly influenced child motivational
variables including self-regulation, perceived competence, and control understanding and
that parent behavior indirectly influenced school grades. Mother behavior was found to
directly influence perceived competence, control understanding, and school grades. In
short, parent actions were found to significantly influence child outcomes directly and
indirectly through child motivational factors.

Research suggests that parent involvement with the schools has an appreciable
effect on the academic lives of children. In particular, Marcon (1999) examined the
relationship between parent involvement and the development and academic performance
of inner-city preschoolers. She used analysis of covariance (ANCOVA) to examine how
level of involvement and type of involvement relate to adaptive behavior, as measured by
the Vineland Adaptive Behavior Scales, and academic performance, assessed through the
district’s Early Childhood Progress Report. Marcon’s results indicated that increased and
more active involvement was significantly associated with improved development and
academics, particularly for preschool boys. Overall, Marcon’s study suggests that parent
involvement plays a large role in development and that special efforts should target the

parents of preschool boys.

49

In summary, three parent outcomes corresponding with the three components of
the Esler model (2002) will be examined. First, parent satisfaction with schools will
represent the attitudes component of the model. Also, parent trust, which represents the
relationship part of the Esler model will be included. Finally, parent involvement
activities, which correspond with actions, will be defined somewhat narrowly as parent-
initiated involvement activities focused on their children’s school (Kohl, Lengua, &
McMahon, 2000). Together, these three measures cover a broad spectrum of parent
outcomes and also tie the current study in the with Esler model of family-school
relationships.

Moderation: Introduction and Specific Moderators

Home-school and social support research suggests that family demographic
information significantly affects the nature of the relationship between the home and the
school (e.g., Malecki & Demaray, 2002; Desimone, 1999; Marcon, 1999). Specifically,
research suggests that student age, parent race/ethnicity and educational attainment, and
school location influences the relationship between home-school collaboration and parent
outcomes through the mechanism of social support. Specific moderators will be
presented below.

Child Grade Level

Child grade level affects the relation between home-school collaboration and
parent outcomes. Specifically, home-school collaboration literature suggests that parent
involvement plays a smaller role as children progress through school, while literature on
social support tells a more complex story about it’s influence as students progress

through school.

50

First, research by 1220, Weissberg, Kasprow, and F endrich (1999) suggests that
parent involvement declines as students move from kindergarten through third grade.
The authors examined teacher perceptions of parent involvement for 1,205 children in
lower elementary school for three consecutive years. Parent involvement was defined as
parent-teacher contact, the parent-teacher relationship, and participation in activities at
home and school. Study results showed that number of parent-teacher contacts, parent
participation at school, and parent participation at home declined significantly over three
years. Despite the decline, results indicated that involvement accounted for a significant
amount of variance in year three educational performance, even after controlling for
initial performance level (1220 et al., 1999). Further, the researchers found that parent
involvement was correlated with student social and academic functioning. Based on their
findings, 1220 et al. suggest that attention be devoted to fostering learning at home
because it predicts the widest range of student outcomes.

Cauce, Felner and Primavera (1982) also studied how social support varies by
child grade. Specifically, they examined social support in high-risk adolescents, teasing
apart the relationship between the structural components of support (family, formal, and
informal support) and the outcomes of academic adjustment and self-concept. Two
hundred and fifty ninth and eleventh graders in the Northeast participated in this study.
Results showed differences in the perceived helpfulness of the support dimensions as a
function of age, sex, and ethnic background, as well as in the relationship of each source
of support and the adjustment indices. In particular, sources of informal support did not
vary by age. Further, for family support, a sex by age interaction was found with

younger males viewing their families as more helpful than older males while Older

51

females rated their families as more helpful than younger females. In terms of formal
support, older adolescents had significantly higher levels of social support than younger
adolescents. Also, older Hispanic adolescents viewed formal supports as more helpful
than younger Hispanic individuals. Based on the findings, the authors recommend that
researchers use multi-dimensional indices of social support and that researchers further
examine how social support varies by race, age, and gender (Cauce et al., 1982).

Taken together, research on child grade level suggests that it plays a varied role in
the lives of students and families. Since the literature does not paint a clear picture, I
intend to base my hypotheses pertaining to grade level on literature on parent
involvement, which suggests that parents have more influence on the lives of younger
children than older children. Thus, I hypothesize that the relationship between the home
and school will have a greater effect on parent outcomes for children in lower elementary
grades (kindergarten through second grade) than for children in upper elementary school.
Family Race/Ethnicity

A second variable that has the potential to moderate the relationship between
home-school collaboration and parent outcomes is family race or ethnicity. As
previously noted, ample research suggests that parent involvement varies by the race and
ethnicity of families (e.g., Desimone, 1999) and that child academic achievement varies
by ethnicity (Anderson & Keith, 1997; Desimone, 1999). Thus, the relationship between
home-school collaboration and parent outcomes will be examined separately by family
race and ethnicity.

For instance, Slaughter and Epps (1987) reviewed a number of qualitative and

survey research studies on the educational attainment of Afiican American students and

52

found that few studies examine class, race and ethnicity. Further, they found that African
American parents affect their children’s academic achievement directly through learning

experiences at home and, “indirectly by their impact on the schools their children attend”
(p. 18). In addition, Slaughter and Epps described the pressing need for more research on
the unique processes that contribute to education for Afiican American children.

In addition, Desimone (1999) suggests that family race affects the relationship
between parent involvement and student achievement. Specifically, Desimone examined
two hypotheses related to parent involvement and child achievement. First, she
hypothesized that parent involvement explains more variance in achievement for White,
Black, and Hispanic students than for Asian students. She also hypothesized that the
parent-school relationships are more strongly related to the achievement of Black and
Hispanic students than Asian or White students. Her results show that her model was a
better predictor for White and Asian students than for Hispanic and Black students. In
addition, her results suggest that parent involvement explains considerably less
achievement for Asian students than for White, Hispanic and Black students.

Finally, Demaray and Malecki (2002) showed that social support also varies by
student race. In particular, they examined child perceptions of overall social support and
support from a parent, teacher, classmate, and close friend. Their study suggests that
Native American students report Significantly less overall social support than children
from other racial and ethnic groups. In addition, Native American students report
significantly less parent, teacher and classmate support than their White, African
American, and Hispanic peers. Also, Afiican American students report significantly

higher parent and teacher support than White students. White students perceive

53

significantly greater teacher support than Hispanic students and significantly greater close
friend support than Native American students. Together, this data suggests that student
perceptions of social support vary dramatically by race and that these relationships should
be examined separately by race.

Overall, these data lead to a number of hypotheses. First, I hypothesize that for
Afiican American, Hispanic and Asian parents, both types of school-provided social
support will have a larger effect on parent outcomes than for Caucasian parents. Further,
for Afiican American, Hispanic, and Asian parents, I believe that instrumental support
will play a larger role than emotional support in parent outcomes. Further, because of
their close identification with the school culture, I believe that emotional support
provided by the schools will have a greater effect than instrumental support on outcomes
for Caucasian parents.

Parent Education Level

A third potential moderator in the relationship between home-school collaboration
and parent outcomes is family education level, which will be assessed by asking parents
to report their highest level of educational attainment. Ample research within the study
of the home-school relationship suggests that parents with lower levels of education tend
to be less involved than parents with more education.

Within the research on collaboration, a precedent exists for using education level
as an indicator of parental disadvantage. In particular, Kohl, Lengua, and McMahon
(2000) define parental risk using parent demographic variables including mother and
father’s educational level, maternal depression, and single-parent status. In a sample of

Caucasian and Afiican American participants, they found that all three components of

54

parent risk level were significantly related to the six dimensions of parental involvement
(Kohl et al., 2000). In particular, parent education was related to parent-teacher contact,
parent involvement in school, teacher perceptions of the parents’ valuing of education,
and parent involvement at home. Based on their findings, they suggest that researchers
further examine school-based opportunities for involvement and that teacher attitudes be
considered in future studies of involvement.

Marcon (1999) also examines how the home-school relationship affects
development for children whose parents are lesser educated. Using an inner-city
preschool sample with high levels of parental disadvantage and low educational
attainment, Marcon found that a higher level of school involvement and more active
types of involvement (like volunteering in the classroom and attending school events) are
associated with more positive development and better academic achievement for
students. Specifically, She found that this sample of preschoolers at high risk due to
reduced SES was greatly influenced by parent involvement activities, even if the
activities were school-initiated or mandated. This study suggests that the relationship
between the home and school are strongly related to child outcomes for a hi gh-risk
population. Marcon cautions that as this study is correlational, determining the direction
of influence between the parent involvement and child outcomes is impossible.

Third, Pettit, Bates, and Dodge (1997) examined whether early supportive
parenting buffers the impact of early family adversity on children by the time they reach
middle school. Their adversity variable included socioeconomic disadvantage (parent
occupation and education level), family stress, and single parenthood. Pettit et al. (1997)

found that “high levels of SP (supportive parenting) mitigated the effects of family

55

adversity on later behavior problems” (p. 908). Overall, this study suggests that family
adversity, including parent education level, definitely affects child outcomes.

Based on the research noted above, I hypothesize that parents with a low level of
educational attainment will benefit more from instrumental and emotional support than
parents with more education. Also, I believe that instrumental support will account for
more variance than emotional support. Conversely, I anticipate that parents with higher
educational attainment will benefit more from school—provided emotional support than
from instrumental support.

School Location

A final variable that has the potential to influence the relationship between home-
school collaboration and parent outcomes is school location. In particular, location refers
to geographic descriptors of a region like urban, rural, and suburban. For the purposes of
this study, location will be measured at the school-wide level, and based on school
neighborhood.

According to Eccles and Harold (1993), family neighborhood incorporates a
variety of factors including neighborhood cohesion, social support, opportunity
structures, norms, dangers, social controls, and role models that all influence parent and
teacher beliefs and practices (p. 571). One challenge with location as a variable is that it
could be confounded by related factors such as family race or educational attainment.
However, the potential benefits of examining location effects on home-school
relationships and parent outcomes are too great to pass up. Writing from a risk and
resilience perspective, Masten, Best, and Garmezy (1990) note that certain ecological

niches provide clear evidence of “cumulative risk” (p. 426). They note that these risks

56

co-occur and potentially increase an individuals’ risk level exponentially (Masten et al.,
1990). In particular, Masten and her colleagues point to differences between families
from low-risk, middle class neighborhoods and families from high-risk, urban
neighborhoods. Based on their review of the literature, they conclude that context always
matters in development and that contexts can foster development strengths like
interpersonal competence and good problem-solving skills. In addition, Masten, Best,
and Garmezy note that schools can mitigate the effects of stressful contexts and can
provide additional support for individual students by modeling problem-solving and
teaching students how to interact appropriately with the world around them.

Thompson (2002) directly examines how school location and neighborhood
disadvantage influences child outcomes in inner city Detroit. His measure of
neighborhood disadvantage included amount of rent, age of neighborhood houses, house
occupancy, and family income. His results indicated that neighborhoods could be
grouped into low, moderate, and high levels of environmental stress (Thompson, 2002).
The inclusion of neighborhood stress resulted in a three percent increase in the prediction
of child achievement for the sixth graders in his study. Thompson argues that despite the
small increase, neighborhood is an important factor for children and that perhaps
different outcome measures would reveal a larger effect.

Research on family location suggests that it makes a difference, particularly when
the context is negative. Thus, I believe that family location will influence which types of
support account for the most variance in parent outcomes. In particular, I believe that
instrumental support will account for the most variance for parents living in urban areas,

and less for parents from rural and suburban areas. Further, I hypothesize that

57

instrumental support will account for somewhat less variance for parents from rural areas
than urban areas. I believe that parents in suburban areas will benefit least from
instrumental support. On the other hand, I believe that emotional support will account for
the most variance for parents living in suburban areas. Also, I believe that emotional
support will account for slightly less variance for parents living in rural areas than
suburban areas. Finally, I hypothesize that emotional support will account for the least
variance for parents living in urban areas.

Summary of the Literature Review

As shown by the preceding review of the literature on the home-school
relationship, this field is marked by theoretical and empirical confusion and by
subsequent gaps within the body of knowledge.

The first gap within the literature is the dearth of empirically supported theories of
the home school relationship. As noted above, many scholars have explored this
phenomena, studying parent involvement as a protective factor, outcomes associated with
home-school collaboration, demographic factors that lead to involvement, and how
schools can foster a closer relationship with families. Despite all of this knowledge, few
researchers have offered a comprehensive theory of the home-school relationship with
the notable exceptions of Epstein (1995), Scott-Jones (1995), and Esler, Godber and
Christenson (2002). In order to address this gap within the literature, the model proposed
by Esler et al. with actions, relationships, and attitudes will be adopted and examined
through a close study of parent outcomes corresponding with attitudes (parent satisfaction

with school), relationships (parent trust), and actions (parent involvement activities).

58

A second gap within the literature and one related to the theoretical tangle is the
fact that research has not Specified a psychological mechanism by which the home-school
relationship influences family outcomes. One potential mechanism with considerable
support within the psychological literature is social support. In this study, emotional and
instrumental forms of social support are the predictor variables and are hypothesized to
influence family outcomes.

A third gap within the literature on home-school relationships is that child
academic outcomes have been studied almost exclusively. This poses a problem for two
reasons. First, because families are proximal to this relationship, the lens should be
broadened to examine how they are affected by their interactions with schools. Second,
while academic outcomes are important, they only capture one facet of education. The
Esler et al. (2002) model suggests that relationships, attitudes, and actions should be
considered as well. Thus, this study examines parent outcomes including parent trust of
schools, parent satisfaction with school, and parent school-oriented activities.

A final gap within the literature is that factors changing the nature of this
relationship including child grade, parent race, parent educational attainment, and school
location have not been examined in regards to parent outcomes. Research indicates that
these variables have the potential to change the relationship between home-school
collaboration and family outcomes. Thus, the relationship between home-school
collaboration and parent outcomes will be examined separately for the parents of children
in kindergarten through second grade and in third through fifth grade. Further, the
relationship will be examined separately for Afiican-American, Caucasian, Asian and

Hispanic parents. In addition, the relationship will be separated for families with low and

59

high levels of educational attainment. Finally, the home-school relationship will be
examined separately for parents of children in urban, suburban, and rural schools.
Research Questions and Hypotheses

After reviewing the available literature, a number of research questions and
hypotheses have emerged. The first question addresses the relationship between home-
school collaboration and parent outcomes. Which type of school-provided social support
(emotional or instrumental) accounts for the most variance in parent satisfaction with
school, sense of trust, and parent involvement activities? The rest of the questions pertain
to potential moderating variables. How does child grade influence the relationship
between home-school collaboration and parent outcomes? Next, does the relationship
between home—school collaboration and parent outcomes differ for Caucasian, Afiican
American, Hispanic and Asian parents? Fourth, how does parent educational attainment
influence the relationship between home-school collaboration and parent outcomes?
Finally, does the relationship between home-school collaboration and parent outcomes
vary by school location? Specific questions and hypotheses follow below.

Question One: Does emotional or instrumental support account for more
variance in parent satisfaction, sense of trust, and parent involvement activities?

I hypothesize that emotional support will account for significantly more variance
in parent satisfaction with school, parent trust, and involvement than instrumental

support.

60

Question Two: How does child grade influence the relationship between home-
school collaboration and parent outcomes?

I hypothesize that the relationship between parent perceptions of school-provided
social support and all parent outcomes will change as a function of child grade. In
particular, I believe that both types of support will have a greater effect on parent
outcomes for children in kindergarten through second grade and a lesser effect on parent
outcomes for children in grades three through five. Stated another way, the parents of
early elementary students will experience more benefit from social support than parents
of upper elementary students.

Question Three: Does the relationship between home-school collaboration and
parent outcomes differ for Caucasian, African American, Hispanic, and Asian parents?

I believe that parent race causes the relationship between social support and
parent outcomes to change. I hypothesize that for Afiican American, Hispanic and Asian
parents, both types of support will account for more variance in parent outcomes than for
Caucasian parents. Further, I believe that instrumental support will play a larger role than
emotional support in parent outcomes for these parents. In addition, I hypothesize that
because of their close identification with the school culture, emotional support provided
by the schools will have a greater effect on parent outcomes than instrumental support for
Caucasian parents.

Question Four: How does parent education influence the relationship between
home-school collaboration and parent outcomes?

I hypothesize that parent educational attainment will cause the relationship

between parent perceptions of school-provided social support and parent outcomes to

61

 

change. In particular, I hypothesize that parents with less education will benefit more
from both types of support than parents with more education. Also, I believe that
instrumental support will account for more variance than emotional support for parents
with less education. Conversely, I anticipate that parents with more education will
benefit more from school-provided emotional support than from instrumental support.

Question Five: Does the relationship between home-school collaboration and
parent outcomes vary by school location?

Finally, I hypothesize that the relationship between school-provided social
support and parent outcomes will change as a function of school location. In particular, I
believe that instrumental support will account for the most variance for parents living in
urban areas, and less for parents from rural and suburban areas. Further, I hypothesize
that instrumental support will account for somewhat less variance for parents from rural
areas than urban areas. I believe that parents in suburban areas will benefit least from
instrumental support. On the other hand, I believe that emotional support will account for
the most variance for parents living in suburban areas. Also, I believe that emotional
support will account for slightly less variance for parents living in rural areas than
suburban areas. Finally, I hypothesize that emotional support will account for the least

variance for parents living in urban areas.

62

Chapter Three: Methodology

This chapter presents the methods used in carrying out the study, giving special
emphasis to the recruitment of participants, data collection, and measure selection. In
particular, the research contexts will be presented, followed by an introduction to the
parent participants. Following that, the procedures used in data collection will be
elucidated, followed by the study measures.

The General Pmective

This study examines how the home-school relationship affects parents and how
that relationship differs by child grade, parent race and ethnicity, parent education level,
and school location. In particular, the purpose of the study is to generalize information
about the perceptions of parents in this sample to the larger population of parents in Mid-
Michigan (Babbie, 1990).

Data was collected using survey methodology. This method was selected for a
variety of reasons. First, anonymous surveys allowed parents to respond candidly about
their perceptions of the schools. Also, families were able to complete the surveys at their
convenience, which potentially increased response rates. Third, survey methodology is a
cost-effective form of research befitting the researcher’s limited funds for data collection
(Creswell, 2003). Also, this method allowed for relatively rapid turnaround in data
collection (Creswell, 2003).

Survey methodology was also appropriate because parent perceptions of social
support were the focus of this study rather than the actual enactment of social support
(Tardy, 1985). Since social support is a phenomenological construct, participant

perceptions are the only way to capture this occurrence.

63

Various researchers have argued that perceptions of support are important and
perhaps even more important than the enactment of support. First, Thoits (1982a) argued
that research on social support should focus on the functional or perceived amount and
adequacy of the support from one’s network. Tardy (1985) also argued that disposition
or availability of support is important separate from its enactment. Disposition refers to
the “quantity or quality of support to which people have access” while enactment refers
to “the actual utilization of those support resources” (p. 188). Third, Demaray and
Malecki (2002) suggest that participant appraisals of support are as important as
enactment. Specifically, they ask participants to rate the amount and types of support as
well as its importance to them in their measure of perceived social support. This suggests
that participant perceptions are highly important in the lives of individuals. In sum,
survey methods are justified in this study for theoretical as well as pragmatic reasons and
participant perceptions are an important part of the research on social support.

The study was comprised of a cross-sectional survey. All of the data collected
during a two-month period to ensure that the home-school relationship was at a similar
point in each school. Cross-sectional research allowed the researcher to capture a
snapshot of the home-school relationship at that point in time.

Research Context

The principals of three public elementary schools in mid-Michigan were
contacted regarding data collection. The principals agreed to participate and to oversee
data collection and the distribution of materials. Please see Table One for the

demographic information on the participating schools and Michigan averages.

64

Table 1

Demographic Information for Participating Schools

 

School l Grades

l
1

Students l Lunch Status* l Racial Profile

l

Achievement **

 

i
9

91% Caucasian

 

 

 

 

 

 

 

 

 

 

 

i Math: 60%
4% African-Am. Reading: 80%
School A K-4 387 13% 3% Hispanic Writing: 31%
, 2% Asian
1% Native Am.
84% Caucasian Math: 73%
l 5% Afiican Am. Reading: 86%
5
School B I 2.5 383 13% 5% Asian Writing: 55%
4% Hispanic
1% Native Am.
i 39% African Am. Math: 44%
33% Hispanic Reading: 48%
School C K-5 205 90% 21% Caucasian Writing: 16%
3% Asian
J 3% Native Am.
74% Caucasian Math: 65%
Michigan 6 18% African Am. Reading: 73%
Average i K-5 36% l 5% Hispanic Writing: 47%
2% Asian

 

 

 

 

2% Native Am.

 

 

"mawd

 

 

*Lunch Status: percentage of students who receive a free or reduced-cost lunch.
”Achievement Data: percentage of fourth graders who meet or exceed state achievement

standards (2003)

Research Participants

Parents are the unit of analysis in this study because I am interested in how parent
perceptions of the home-school relationship influence parent outcomes rather than school
practices for improving the relationship. Since parents are the most proximal part of this
relationship, they are the most logical source of data. Further, as previously noted, parent
outcomes have largely been ignored within the literature on the home-school relationship.

In order to attain a 95% confidence interval, an alpha level of .05, a power level
of .80, and the ability to detect a small effect size of .25, I attempted to recruit at least 199
parents (Howell, 2002). Potential sample sizes range from 199 (the most liberal) to 346
if I wanted a more conservative sample size and a greater ability to detect true significant
differences.

Two hundred and fifty parents returned surveys and consent. One hundred and
twenty one parents from school A, 73 parents from school B, and 56 parents from school
C participated. 83.1% of the responding parents were mothers. Most parents who
returned surveys were Caucasian although parents from a range of racial and ethnic
groups also participated. In particular, 73.7% (n = 182) of the sample was Caucasian,
10.2% (n = 26) was Hispanic, 7.1% (n = 18) was Asian, 6.1% (n = 15) was African

American, and 2.4% (n = 6) selected “other” as their race/ethnicity and tended to specify

66

being multi-racial and Native American. Two point eight percent (n = 7) did not report
their race.

Participating parents also varied in educational level. Zero point four percent of
parents (n =1) completed no high school, 9.4% (n = 24) of parents completed some high
school, and 14.6% (n = 37) completed either high school or a GED. Eighteen point nine
percent (n = 48) completed some college, 9.4% (24) attained an Associates degree, 22.8%
(n = 58) completed a Bachelors degree, 5.9% (n = 15) completed some post-college
courses, and 15.4% (n = 39) completed a graduate degree. Three point one percent (n =
8) did not report their education levels.

Participating parents also provided demographic information about their children.
Forty-nine point eight percent of their children were boys and 20.4% of the children
received special education services. The students’ grades ranged from kindergarten to
fifth grade. Nineteen point six percent of the students were in kindergarten (n = 48),
19.6% were in first grade (n = 48), 18.0% were in second grade (n = 44), 18.0% were in
third grade (n = 44), 13.9% were in fourth grade (n = 34), and 11.0% (n = 27) were in
fifth grade. The decrease in parent responses is consistent with literature on parent
involvement across elementary school. In addition, this was compounded by the fact that
the largest school, School A, only included kindergarten through fourth graders. Please

see Table Two for information regarding the participating schools.

67

Table 2

Demographic Information for Participating Parents

 

 

 

 

 

 

 

 

 

Parent Percent ' Parent Racial Profile of
School
Participants Participating Education* Participants"
87% Caucasian
School A 5% Hispanic
5.5% Less
Suburban i 73 19% . 5% Asian
1 E 94.5% More
Public E 3% African American
1 l
i E 1% Other
l
92% Caucasian
School B E E
15.0% Less 3% Hispanic
Suburban 121 31% E
, 85.0% More 3% Asian
Public l
E 3% oma
33% Hispanic
1
School C E 22% Afiican American
75.5% Less
Urban 56 27% 22% Caucasian
24.5% More
Public 18% Asian

 

 

 

l
E

 

6% Other

 

* Less Education includes the categories: No High? School, Some High School, GED, or a High

School Diploma. More Education includes: Some College, Associates Degree, Bachelors

Degree, or an Advanced Degree.

** Percentages may not add up to 100% because of rounding

68

 

Procedures

Data collection included a number of steps and procedures. Steps included pilot
testing the study survey and revising the newly created Parent Perceptions of School-
Provided Social Support (PPSSS). Further steps included gaining principal consent and
University Committee on Research Involving Human Subjects (UCRIHS) approval.
Then, data collection commenced followed by data entry into an SPSS data set. Finally,
incoming data were checked for response bias using wave analysis.

First, the study survey including the PPSSS, was pilot tested with ten parents.
These parents were selected because they were accessible to the researcher and because
their children attended schools that were similar to potential data collection sites. Parents
were asked to complete the surveys and also to rate the surveys for clarity. In particular,
they were asked to respond to each PPSSS item with the questions, “Is this question
clear? Does it make sense?” Parents were asked to respond by selecting, “No,”
“Somewhat,” and “Yes.” In addition, parents were asked to provide written feedback on
the readability, visual interest, wording, clarity, and length of the survey. One parent
noted that some items were not equally applicable to all parents because of poverty and
education. Another parent suggested making the response choices more consistent. This
change was made prior to actual data collection.

PPSSS items with low clarity ratings (at or below a response two-thirds of the
way between “Somewhat” and “Yes”) were examined. Five of the 24 items on the scale
had clarity ratings below the cutoff. The lowest clarity item had a score of 2.44. Two of
the five items were deemed unclear by the researcher and her advisor, and thus were

reworded. The item, “My child’s school lets me know that the things I do help my child

69

learn” was replaced with “My child’s school thinks parents are important for learning.”
Also, the item, “My child’s school tells me how I’m doing compared to other children in
my child’s classroom” was replaced with, “My child’s school has clear expectations for
my participation in my child’s learning.”

Prior to rewording the items, the reliability of the original PPSSS and its subscales
was calculated. The Cronbach’s alpha of the original PPSSS was 0.77. However, this is
likely a low estimate because it only included the five cases with complete data. The
Socioernotional scale had an alpha value of .94 and the Instrumental Scale had a value of
.78. Both alpha values were sufficiently high for data collection to commence (Howell,
20002). Actual data collection began with the revised PPSSS featuring the two reworded
items.

Following pilot testing, three elementary school principals were contacted
regarding participation. These principals were selected because of their willingness to
engage in research and because of their relationships with the students and faculty of the
School Psychology program at Michigan State University. After a telephone
conversation, the principal for school A agreed to participate in the research. The
principal for school B asked that an application to conduct research be completed with
the assistant superintendent of the district. Following approval from the assistant
superintendent and a meeting, the principal at school B agreed to participate in the
research. Both principals completed letters of intent to participate for inclusion with the
UCRIHS application. The principal at school C was enthusiastic about participating but
needed approval from the school board prior to participation. After obtaining permission

from his school system, the principal at school C gave permission.

70

 

Following pilot testing and principal assent, University Committee on Research
Involving Human Subjects (UCRIHS) approval was sought. The UCRIHS committee
approved the study survey as part of a larger study, One Pastor One School, on February
23, 2005. The reviewers did not suggest any revisions to the original documents.
Following initial approval, the researcher revised the consent form and changed some of
the items of the study survey. Re-approval was granted on March 28, 2005.

After final UCRIHS approval, the introductory letters for parents and teachers
were dropped off in the school offices at schools A and B. Please see Appendices One
and Two for the complete documents. One week later, the Home-School Relationship
Survey packets were dropped off at schools A and B for distribution by the classroom
teachers. Included in each parent packet was the study survey (Appendix Three), the
consent form (Appendix Four), and a postage-paid mailing envelope. The study survey
included an introductory paragraph that explained the study and gave parents directions
for completion and for returning the forms. Two weeks later, a reminder letter was sent
home with parents (Appendix Five). Three weeks after the reminder letters were sent
home, a second set of survey packets was sent home to the parents at school A who did
not return the initial surveys. The principal for school B opted not to send out a second
wave of surveys because she was participating in another research project and did not
want to overwhelm staff and parents. The principal at school C also thought it best to
distribute one wave of surveys rather than two because of time constraints pertaining to
the close of the school year.

Parents completed and returned the surveys and consent forms. Parents at schools

A and B returned the surveys to the researcher via postage-paid envelopes. Parents from

71

school C returned the surveys in sealed manila envelopes to their school office because of
time constraints and in order to increase parent participation.

In order to avoid multiple data points from the same family and a potential
reduction in variance, two requests were made of parents. First, only one parent was
asked to complete the survey. Second, parents were asked to complete the survey for
only one child per elementary building. In the case of multiple students from the same
family within a school, parents were asked to complete the survey on the youngest child
in kindergarten through fifth grade.

Data from parents was entered into an SPSS dataset. This statistical program was
selected because it offers ample flexibility in data analysis and because it is a user-
friendly program. It is also well equipped to handle large datasets consisting of many
variables and participants.

Wave analysis was also conducted to test for response bias. Response bias occurs
when participants who respond earlier to surveys have more positive views than
participants who respond later in data collection. In particular, wave analysis involves
examining the returns on three randomly selected items week by week to determine if the
average response changes over time (Creswell, 2003). Three questions were selected
using a random number table to select items. The three questions selected were, “My
child’s school gives me advice about my child’s learning and development,” “I trust that
my child’s teachers are doing a good job encouraging my child to have a positive attitude
towards learning,” and “I trust that teachers have my child’s best interests at heart.”
Then, a crosstabs function was used to display the three variables with scores by date.

Next, the dates were grouped into weeks one through seven. The means for each week

72

were then calculated and are presented below. Please see Figure One for complete
details. Visual analysis reveals no or slight downward trends in parent responses to the
three questions.

Figure 1

Wave Analysis Results for Parent Responses Over Time

 

Wave Analysis of Three Items

 

 

6
o
8
g - _-_ -7”
a +Advice
g 3 i» —— --— — e - — i +Attitude
’E 2 L, L . _ - L _ Interestsl
8 l "T ’ ““T
a ,, e a ~ + c

1 2 3 4 5 6 7
Time: Weeks 1 to 7

 

 

 

Measures Used in DatIaCollection

Four measures were used in the data collection process. The predictor variable
was a newly created measure called the Parent Perceptions of School-Provided Social
Support (PPSSS). The outcome variables were the Trust Scale from the Family School
Relationship Survey (Adams & Christenson, 2000), a composite parent satisfaction with
school scale, and the Parent Report of Involvement from the Parent Teacher Involvement
Questionnaire (Kohl et al., 200). The moderator variables were Child Grade, Parent
Race/Ethnicity, Parent Educational Attainment, and School Location. Please see
Appendix Three and the complete Home-School Relationship Survey for the complete

scales and items.

73

Predictor Variable
Parent Perceptions of School-Provided Social Support

A new measure was created to assess parent perceptions of school-provided social
support. It was created by modifying Malecki and Demaray’s (2002) CASSS (Child and
Adolescent Social Support Scale). Author permission was obtained for the use and
adaptation of the existing CASSS. The newly created scale is called Parent Perceptions
of School-Provided Social Support (PPSSS). This scale retains the four categories of
emotional, appraisal, instrumental and informational support initially specified by House
(1981). Some items from Malecki and Demaray’s CASSS were retained and modified
for use in the PPSSS.

In accordance with social support theory, a socioemotional support and an
instrumental support subscale were developed and each subscale was examined
separately as a predictor. Parents were asked to respond to 12 questions about emotional
support, which included appraisal support, and 12 questions about their perceptions of
instrumental support, including informational support, from their child’s school. An
example of emotional support includes, “My child’s school understands my child” and an
example of instrumental support is “My child’s school helps me find resources (like
information about after-school programs or summer camps)” Parents responded on a
five-point Likert—type scale with one corresponding with “Not at All” and five with
“Definitely.” As noted above, the PPSSS was pilot tested with a small group of parents

prior to wide-scale administration.

74

Outcome Variables
Parent Trust of School

Adams and Christenson’s Trust Scale from the F amily-School Relationship
Survey (2000) was used to assess parent trust of school. Author permission was obtained
for the use and potential modification of this scale. The scale was used as written except
for two modifications. First, the stem for the questions was changed fiom “I am
confident that teachers...” to “I trust that teachers...” to more explicitly address the idea
of parent trust. Second, the response choices were changed to reflect a five-point Likert-
type scale rather than a four-point Likert-type scale. The original scale was developed to
reflect basic components of trust including predictability, dependability, and faith as
specified by Rempel, Holmes, and Zanna (I 985). Sample items from this scale include “I
trust that teachers respect me as a competent parent,” “I trust that teachers are doing a
good job teaching my child academic subjects,” and “I trust that teachers are fiiendly and
approachable” (Adams & Christenson, 2000; p. 485). Parents were asked to respond to
these items by marking a response from one to five.
Parent Satisfaction with School

Parent satisfaction with school was assessed using a newly created parent
satisfaction with school scale. The satisfaction scale was comprised of a global
assessment of satisfaction (Kohl et al., 2000) and parent satisfaction with specific
domains of school (Goldring & Shapira, 1993). The measure from Kohl and colleagues
was the Parent Endorsement of School (PES) subscale from the Parent Teacher
Involvement Questionnaire (PTIQ). The domain-specific measure was an adaptation of

Goldring and Shapira’s eight-item scale.

75

Permission has been Obtained for the use and adaptation of Kohl, Lengua, and
McMahon’s (2000) PTIQ in this study. The PES subscale was used as written with slight
modifications in wording to increase parent comprehension of the questions. The
original scale includes four items tapping into parent’s affective appraisal of the school.
Items within the study survey include, “School staff is doing good things for my child”
and “I have confidence in the people at my school.” Other items address the school’s fit
for the child and how the school is preparing the child for the future. As with the original
scale, parents were asked to rate their agreement with the items on a five-point Likert
type scale and again response options again ranged from “Not at All” to “Definitely.”

The second measure of satisfaction was loosely adapted from Goldring and
Shapira’s (1993) article, “Choice, Empowerment, and Involvement: What Satisfies
Parents?” The authors examined parent satisfaction with eight areas of school
functioning including, “academic, social, citizenship, general school atmosphere, values,
educational philosophy, developing individual potential, and the curriculum” (p. 403).
The original satisfaction scale was composed in Hebrew without an English translation
and thus the original measure was not used for this study. However, Ellen Goldring
granted the researcher permission to create new satisfaction items based on the eight
categories of satisfaction. Sample items include, “How satisfied are you with the
school’s promotion of your values and beliefs?” and “How satisfied are you with the
school’s friendliness?” The global satisfaction scale and the domain specific scales were

combined to create one l2-item measure of satisfaction.

76

Parent Report of Involvement Activities

The parent involvement scale was an adaptation of Kohl, Lengua, and
McMahon’s (2000) measure of involvement from the parent form of the PTIQ. As noted
above, permission has been obtained for the use and adaptation of this measure. The
Conduct Problems Prevention Research Group originally developed this questionnaire,
which is based on Grolnick and Slowiaczek’s (1994) conceptualization of involvement,
which hypothesizes that parent behavior at school and at home influences children’s
academic achievement.

The existing measure was adapted in a number of ways. First, only parent
perceptions of the relationship were assessed; teacher perceptions were not examined
since they have already been the focus of numerous studies (i.e., 1220, et al., 1999). Also,
only selected subscales of the PTIQ were included because of potential overlap with the
trust and satisfaction measures. In addition, the item format was revised.

Subscales included in the present study included: Parent-Teacher Contact, Parent
Involvement at School, and Parent Involvement at Home. Parent-Teacher Contact
examined parents’ relationship building with their child’s teacher. Ideas in this section
included calling a teacher and attending parent-teacher conferences (Kohl et al., 2000, p.
512). The Parent Involvement at School subscale included volunteering at school and
visiting “school for special events” (p. 512). Parent Involvement at Home included the
questions, “Do you read to your child?” and “Do you take your child to the library” (p.
513)? All of these items were put into the present tense with the stem, “How often do

you...” For example, one item from the study survey is “How often do you write notes

77

or email to your child’s teacher?” Parents were asked to respond to these 11 questions
using a five-point Likert-type scale with options ranging from “Never” to “All the Time.”
Moderators

Demographic information pertaining to child grade level, parent race or ethnicity,
parent education level, and school location were collected from the participating parents.
Information related to parent gender, child gender, and special education services was
also collected.
Child Grade

Participants indicated their child’s grade level by responding to the question, “In
what grade is your youngest child in this building?”
Family Race

Parent race was assessed. Parents were asked, “What is your race/ethnicity?”
firent Education Level

Parent education level was assessed using the following item: “Please circle your
highest educational level.”
School Location

School location was assigned to schools based on their geographic location. Two
university-affiliated raters were asked to assign schools to one of the three categories
(urban, rural and suburban). The raters had 100% congruence in categorizing the three
schools.
Summary of the Methodology

Survey methodology was used to examine how the home-school relationship

influences parent outcomes through the mechanism of social support. First, pilot testing

78

was conducted and UCRIHS approval was sought and granted. Following scale
development and UCRIHS approval, 250 parents were recruited from an urban and two
suburban schools. The independent variables were the newly created socioemotional and
instrumental subscales of the PPSSS. The dependent variables were scales addressing
parent trust of school, satisfaction with school, and parent involvement activities. Also,
information on child grade, parent race, parent education level, and school location was

collected to study potential moderation effects.

79

Chapter Four: Results
This chapter included two main sections. In the first section, the psychometric
properties of the measures were examined, data transformation procedures were detailed,
and descriptive statistics were analyzed. In the second section, the five hypotheses were
examined using correlation and multiple regression analyses. The hypotheses were
organized by the potential moderators of child grade, parent race, parent education, and
school location.
Measure Preparation
Psychometric Properties of the Measure
The first aim of this study was the creation and validation of the Parent
Perceptions of School-Provided Social Support scale (PPSSS). The PPSSS was validated
in two ways. First, Pearson correlation coefficients were calculated for the scale items.
This procedure determines the strength and direction of linear relationships between two
variables. The 24 items were significantly and somewhat highly intercorrelated with
Pearson correlation values ranging from .52 to .84. Confirmatory factor analysis was also
performed. Factor analysis uses a mathematical model to account for the most variance
in a pattern of correlations and is used to group related variables. A principal component
factor analysis procedure with Varimax rotation was used to examine the psychometric
properties of the scale. The results confirmed a strong unitary factor with only one
eigenvalue over one (16.80), which accounted for 69.99% of the variance. All items
loaded onto a single factor (all factor loadings above 0.532). The Cronbach’s alpha value

for this scale was .98.

80

Consistent with research and theory on social support, original plans included
examining the socioemotional and instrumental support as separate predictors.
Unfortunately, the items in the two scales were too highly correlated (.84) for separate
analyses. Thus, the PPSSS was used in its entirety as the predictor.

The second new measure was the aggregate satisfaction scale including global
and domain-specific satisfaction. The 12 items were somewhat highly intercorrelated
with Pearson correlation values ranging from .60 to .91. The validity of the scale was
also explored using a principal component factor analysis procedure with Varimax
rotation. The results confirmed a strong unitary factor with only one eigenvalue over one
(9.01), which accounted for 75.08% of the variance. All items loaded onto a single factor
(all factor loadings above 0.795). The Cronbach’s alpha value for this scale was .97.

In addition to the PPSSS and the satisfaction measures, the Cronbach’s alpha
values for the other, existing scales were computed. Since these scales had already been

validated within the literature, another factor analysis was not performed. Please see

Table Three below for details.

81

Table 3

Summary of Reliability for Existing Measures

 

 

 

 

 

 

Construct Title Items Published a Current (1
Trust Family School Relationship Survey: 19 .96 .98
Trust Scale

Satisfaction: PTIQ: Parent Endorsement of 4 .92 .97
Global School subscale

Satisfaction: Satisfaction in specific 8 .88 .95
Specific school domains

Parent PTIQ: Parent Teacher Contact; l l .67 to .89 .77

Involvement Parent Involvement at School; Parent

Involvement at Home

 

Data Transformation

Following scale development, the data were visually and statistically examined.

All but the parent involvement scale revealed significant negative skewness and kurtosis.

The skewness values were -1 .55 (SE = .15) for social support, -2.06 (SE = .15) for parent

trust, and -1.58 (SE = .15) for parent satisfaction with school. Kurtosis values were also

computed prior to transformation. Kurtosis scores for social support were 2.19 (SE =

.31); scores for trust were 4.33 (SE = .31), and scores for satisfaction were 2.02 (SE =

.31). In addition, the Kolmogorov-Smimov values and Shapiro-Wilk values for support,

trust, and satisfaction were all significant, which indicated a lack of normality.

Because the data were significantly negatively skewed and kurtotic, logarithmic

transformations were computed using the SPSS algorithm on the social support scale, the

82

 

parent trust scale, and the parent satisfaction scale. In addition, the predictor and three
parent outcomes were centered to prepare for the regression analyses. Also, when data
were missing, the participant was excluded on a case-wise basis.

Throughout the text, presented results account for the change in direction
attributable to the logarithmic transformation of the social support, trust, and satisfaction
variables.

Another issue emerged during the initial inspection of the data. Following data
entry, it became clear that significant co-vari ation existed between race and school
location. In particular, the two suburban schools were primarily comprised of Caucasian
families. The third school has large Hispanic, African American, and Asian populations.
Because parents were largely separated by race between schools, it was impossible to
separate race from school location in analyses. Thus, the question of race was not
examined separately in this study. For further discussion of this decision, please see the
hypothesis pertaining to the potential moderator of parent race.

Descriptive Statistics

Descriptive statistics were computed for the newly transformed measures. The
means, standard deviations, skewness, and kurtosis were computed. In addition, the data
was visually examined; a normal curve was overlaid over each scale’s distribution to

check for normality. Please see Table Four for details.

83

Table 4
Descriptive Statistics for Social Support, Parent Outcomes, and Moderators Prior to
Transformation

Scale Standard

Scale Items Mean Skewness Kurtosis
Mean Deviation

 

Social Support

Total 99.79 22.04 24 4.16 -l.55 2.19

 

Parent Outcomes

Trust 82.86 16.86 19 4.36 -2.06 4.33
Satisfaction 52.41 9.68 12 4.37 -1.58 2.02
Involvement 38.46 6.42 1 1 3.50 -0.1 1 -O.17

 

84

Table 5

Descriptive Statistics for Social Support, Parent Outcomes, and Moderators After
Transformation

Skewness Kurtosis

Scale Standard

Mean Deviation

 

 

 

Social Support a 0.00 (1.00) -0.17 -1.13
Parent Outcomes
Trust 3" 0.00 0.61 0.07 -135
Satisfaction 8" 0.00 0.53 0.1 l -1.34
Involvement” 0.00 6.42 -O.11 -0.l7
Moderators
Child Grade C 0.41 0.49 0.35 -1.89
Parent Education C 0.54 0.50 -0.14 -2.00
School Location c 0.76 0.43 —1.25 -0.44

 

Note. Parent race was excluded from this table because it was not used in the final
analyses.
3 Variable was centered.
b Variable was logarithmically transformed.
° Variable was dummy coded.
Data Analysis Strategy
The data were examined using a variety of statistical techniques. First, the item
means were examined prior to transformation. In addition, a correlation matrix was

constructed to examine the strength of the relations between social support, parent

85

 

outcomes, and the interactions. Also, multiple regression was used to examine to the
relative contribution of social support to the three parent outcomes and to test for
moderation by child grade, parent race, parent education, and school location.

The first method used to examine the data was a visual analysis of the item means
prior to transformation. The item means indicated that parents felt supported by their
child’s schools. The average social support score was a 4.16 out of five, which
corresponds with “I agree mostly” with most of the social support items. Further, the
socioemotional support scale had an average value of 4.22 out of five, while the
instrumental scale had an average response of 4.12 out of five. A paired samples t-test
was used to determine that parents perceive receiving significantly more socioemotional
support than instrumental support (t(247) = 3.87; p = .000). However, this difference is
likely a result of the study’s sample size rather than a practically significant difference.

The parent outcome variables revealed high levels of parent trust, satisfaction, and
involvement. Parents rated their trust of their child’s school as 4.36 out of five, which
con'esponds with “I agree mostly” with the trust items. The satisfaction average score
was 4.37 out of five. The parents reported generally lower levels of involvement and
their average involvement score was a 3.50 out of five. However, this is encouraging
because it suggests that parents did not display a positive response bias, selecting mostly
high marks in rating themselves and their child’s schools.

The second method used to analyze the data and to test the hypotheses was
correlation analysis. Bivariate correlations were calculated between the continuous
variables of social support, trust, satisfaction, and involvement. Point—biserial

correlations were calculated between the categorical dummy-coded moderators of grade,

86

education, school location, and the interactions of those moderators with social support.

Please see Table Six for details. Only correlations that were significant at the .01 level

were examined in order to maintain a family-wise error rate of .05.

Table 6
Correlations Between Measures

Measure 1 2 3 4

 

5 6 7
1. Social
-- .085 .025 -.028 .797** .759** -.151*
Support ab
2. Grade -- .125* .072 .118 .117 -.068
3. Education -- -.523** 0.36 0.48 .365"
4. Location -- .009 -0.25 .351 **
5. Parent
-- .788" -.161*
Trust ab
6. Parent
—- -.091

Satisfaction ab

7. Parent

Involvementab

 

aVariable was centered.
bVariable was logarithmically transformed.
*p<05

**p<.01

87

Some of these correlations bear mentioning yet were not addressed by specific
hypotheses. First, the moderators of parent education and school location were strongly
positively correlated (r = .52, p <.001) with lower levels of education associated with
urban schools. Parent education was also correlated with parent involvement activities (r
= .37, p <.001). Again, higher levels of education were correlated with higher levels of
involvement. Further, school location was correlated with parent involvement with
parents with parents at suburban schools demonstrating more involvement than those at
urban schools. Finally, the outcomes of parent trust and satisfaction with school were
also strongly positively correlated (r = .79, p <.001) with higher levels of trust associated
with higher levels of satisfaction.

Following an examination of the correlations and prior to conducting regression
analyses, statistical strategies were employed to test the assumptions for regression
analyses. The four main assumptions in multiple regression are adequate sample size,
low multicollinearity, the exclusion of outliers when necessary, and an adequate
distribution related to normality, linearity, homoscedasticity, and independence of
residuals (Pallant, 2001).

Sample size was calculated using Tabachnick and F iddell’s formula (1996, p.
132). The formula is: N > 50 + 8m (m = number of independent variables). Thus, I
needed at least 50 + 8(4) or 82 participants and my sample size was 250 parents.

The issue of multicollinearity was also addressed. As noted above, the social
support subscales were combined because of high multicollinearity. Another potential
challenge is the relatively high correlations between social support and parent trust, and

parent trust and parent satisfaction. More conservative authors suggest that a correlation

88

about .7 might be problematic (Tabachnik & F iddell, 1996) but others suggest that
multiple regression is relatively robust to violations of these assumptions (Howell, 2002).
In addition, the Tolerance column under the Collinearity Statistics had a value of 1.00.
Since low values (near zero) indicate potential collinearity, the multicollinearity was
likely not problematic (Pallant, 2001).

Outliers were examined by inspecting the residuals scatterplot and Mahalanobis
distance values. A visual inspection did not reveal any outliers. Outliers were also tested
using the Mahalanobis distances. A chi-square critical value of 7.88 was used, which
was determined by using a chi-square distribution table and an alpha of .005 with one
degree of freedom because there was one independent variable (See Pallant, 2001 for
more information). Since the most extreme distance value was 3.28, well below the cut-
off for problematic values of 7.88, outliers were not a concern (Pallant, 2001).

Fourth, the distribution was examined for normality, linearity, homoscedasticity,
and independence of residuals through a review of the normal probability plot and the
residuals scatterplot. The points in the normal probability plot lay in a relatively straight
diagonal line with a relatively even cigar shape from bottom left to top right which
suggests normality, linearity, and homoscedasticity. In addition, the scatterplot of the _
residuals suggested that the residuals were relatively linear, independent, and evenly

distributed about predicted dependent variable scores.

89

Examination of Hypotheses

Question One: Does socioemotional or instrumental support account for more
variance in parent satisfaction with school, sense of trust, and parent involvement
activities? I hypothesized that socioemotional support would account for significantly
more variance in the three parent outcomes than instrumental support.

Due to the high multicollinearity (r = .92, p <.001) between the centered and
logarithmically transformed socioemotional and instrumental support subscales, this
question could not be answered. This high correlation between the variables suggests
that dividing social support into four types or even two larger types is not statistically
appropriate with the parents of elementary school students. As evidenced by the high
means in parent reports of social support, parent perceptions of school-provided social
support are important, but the particular type of support is not as relevant. In summary,
parents certainly perceive receiving support but do not necessarily differentiate between
the various types.

Instead, the socioemotional and instrumental scales were combined to make one
social support variable accounting for emotional, appraisal, informational, and
instrumental support. This combined variable acted as the predictor in the multiple
regression equations with trust, satisfaction, and parent involvement as the outcome
measures.

Social support accounts for a large amount of variance in parent trust and
satisfaction and a much smaller amount of variance in parent involvement. Social

support explains 63.5 % of the variance in parent trust (,8 = .80, F(l , 247) = 427.38, p <

90

.000). Social support also accounts for 57.7% of the variance in satisfaction (fl = .76,

F (1, 247) = 335.27, p <.000). Social support accounts for only 2.3% of the variance in
parent involvement (,8 = .15, F (1, 244) = 5.66, p =.018). In short, social support accounts
for the most variance in parent trust, followed closely by satisfaction and then by
involvement.

Question Two: How does child age influence the relationship between home-
school collaboration and parent outcomes?

I hypothesized that the relationship between parent perceptions of school-
provided social support and the three parent outcomes would change as a function of
child grade. Specifically, Ibelieved that socioemotional and instrumental support would
have a greater effect on parent outcomes for children in kindergarten through second
grade and a lesser effect on parent outcomes for children in grades three through five.
The combination of socioemotional and instrumental support will not influence this
hypothesis since I did not specify which type of support would have more effect.

In order to examine the relationship between social support and the parent
outcomes as a function of child age, regression analyses were run. An interaction term
was created by multiplying the dummy code for child grade (with kindergarten through
second grade as the reference group) by the social support variable (Aiken & West,
1991). This term was entered following social support and the dummy code for child
grade (Baron & Kenny, 1986). Regression results indicated that the interaction term was
non-significant for all three parent outcomes. The standardized regression coefficient for
trust was .06 (t = 1.07, p = .29), the standardized regression coefficient for satisfaction

was .03 (t = .58, p = .56) and the standardized regression coefficient for involvement was

91

-.04 (t = -.41, p = .68). In summary, social support does not play a different role in the
parent outcomes for parents of younger children than for the parents of upper elementary
students. Please see Table Seven for further details.

Table 7

Summary of Hierarchical Regression Analysis for Social Support and Child Grade in

Predicting Parent Trust, Parent Satisfaction, and Parent Involvement

 

 

 

 

 

Parent Trust F Sig. AR2 B B T Sig.
Step 1 211.98 .00 .64

Social Support .48 .79 20.39 .00

Grade .06 .05 1.25 .21
Step 2 141.78 .00 .00

Social Support .46 .75 14.12 .00

Grade .06 .05 1 .23 .22

Social Support x Grade .05 .06 1.07 .29
Parent Satisfaction F Sig. AR2 B [3 T Sig.
Step 1 166.13 .00 .58

Social Support .40 .75 18.04 .00

Grade .05 .05 1.17 .24
Step 2 110.57 .00 .00

Social Support .39 .73 12.72 .00

Grade .05 .05 1.16 .25

Social Support x Grade .03 .03 .58 .56

 

92

Parent Involvement F Sig. AR2 B [3 T Sig.

 

 

Step I 3.61 .03 .03
Social Support 1.0] .16 .25 .02
Grade .74 .06 .90 .37
Step 2 2.45 .06 .00
Social Support .85 .13 1.50 .13
Grade .85 .13 1.50 .38
Social Support x Grade .34 .04 .41 .68

 

Question Three: Does the relationship between home-school collaboration and
parent outcomes difler for Caucasian, African American, Hispanic and Asian parents?

I hypothesized that parent race would cause the relationship between social
support and parent outcomes to change. In particular, I hypothesized that for African
American, Hispanic and Asian parents, both types of school-provided social support
would have a larger effect on parent outcomes than for Caucasian parents. Further, I
believed that instrumental support would play a larger role than emotional support in
parent outcomes. Also, I hypothesized that because of their close identification with the
school culture, socioemotional support provided by the schools would have a greater
effect on parent outcomes than instrumental support for Caucasian parents.

As noted above, socioemotional and instrumental support were combined because
of high multicollinearity so that part of the question cannot be answered. However, the

issue of for which parents perceptions of support play a larger role remains.

93

However, this question could not be answered either. Racial or ethnic status and
school location were confounded because of the three participating schools. Specifically,
schools A and B were predominately Caucasian while school C was comprised primarily
of Hispanic and Afiican American parents. Thus, it was impossible to separate the
effects of parent race and ethnicity from school location. My data reflect differences in
both race and education level. Because educational attaimnent is more closely aligned
with the school-related outcomes in this study, educational attainment was examined in
this study and the analyses of parent race will be dropped. Further, less research has been
conducted on the importance of school location than on race or educational level.

Question Four: How does parent education level influence the relationship
between home-school collaboration and parent outcomes?

I hypothesized that parent education level would cause the relationship between
social support and parent outcomes to change. In particular, I believed that parents with
less education would benefit more from social support than parents with more education.
Also, I believed that instrumental support would account for more variance than
socioemotional support. Conversely, I anticipated that parents with more education
would benefit more from school-provided socioemotional support than fi'om instrumental
support.

With regards to this hypothesis, I was unable to determine which types of support
were more important for parents because of combining socioemotional and instrumental
support. On the other hand, I was still able to determine how parent education influenced

the parent trust, satisfaction, and involvement.

94

Regression analyses suggested that parent education moderated the relationship
between social support and parent satisfaction. First, an interaction term was created by
multiplying the dummy code for parent education (parents whose highest completed
degree was a high school diploma/GED or below comprised the reference group) by the
social support variable (Aiken & West, 1991). This term was entered following the
social support variable and the dummy code for education level. Presented values reflect
the true relationship between the variables, accounting for differences in direction related
to centering and logarithmically transforming certain variables.

Regression results indicated that the interaction term was significant for parent
satisfaction (standardized ,8 = .146, t = 2.47, p = .01). Because the interaction term was
significant, regression equations were plotted for parents with less education and more
education (Aiken & West, 1991). Low, mean, and high average values of social support
were plotted for both groups of parents. Parents with less education experienced more
satisfaction as a result of high social support than parents with more education. At the
same time, when parents perceived low levels of social support, those with less education
experienced less satisfaction than parents with more education. Please see Table Eight
for further details. Please see Figure Two for a graphical representation of the
interaction.

The results indicated that parent education does not moderate the relationship
between social support and parent trust and parent involvement; the standardized
regression coefficient for trust was .02 (t = .36, p = .72) and the standardized regression

coefficient for involvement was .02 (t = .29, p = .77).

95

Table 8
Summary of Hierarchical Regression Analysis for Social Support and Parent Education

in Predicting Parent Trust, Parent Satisfaction, and Parent Involvement

 

 

 

 

 

Parent Trust F Sig. AR2 B 8 T Sig.
Step 1 210.19 .00 .63

Social Support .48 .80 20.49 .00

Education .02 .02 .49 .62
Step 2 139.67 .00 .00

Social Support .49 .81 14.60 .00

Education .02 .02 .49 .62

Social Support x Education .02 .02 .36 .72
Parent Involvement F Sig. AR2 B [3 T Sig.
Step 1 22.72 .00 .16

Social Support 1.10 .17 2.90 .00

Education 4.69 .37 6.17 .00
Step 2 15.12 .00 .00

Social Support 1.21 .19 2.25 .03

Education 4.69 .36 6.1 5 .00

Social Support x Education .22 .02 .29 .77

 

96

Parent Satisfaction F Sig. AR2 B B T Sig.

 

 

 

Step 1 165.28 .00 .58
Social Support .40 .76 18.15 .00
Education .04 .03 .8 1 .42
Step 2 114.53 .00 .01
Social Support .45 .86 14.62 .00
Education .04 .03 .81 .42
Social Support x Education .11 .15 2.47 .01
I Figure 2

Social Support and Parent Satisfaction as Moderated by Parent Education

 

 

Social Support and Parent Satisfaction as
Moderated by Parent Education

 

 

 

 

E-II-Less Education 7;

E-I-More Education .

 

 

 

Parent Satisfaction

 

 

Low, Medium, and High Levels of
Social Support

Question Five: Does the relationship between home-school collaboration and
parent outcomes vary by school location?
Finally, I hypothesized that the relationship between school-provided social

support and parent outcomes changed as a function of school location. In particular, I

97

believed that instrumental support would account for the most variance for parents living
in urban areas, somewhat less variance for parents in rural areas, and the least for parents
from suburban areas. On the other hand, I believed that socioemotional support would
account for the most variance for parents living in suburban areas, followed by rural areas
and then urban areas.

As noted above, race/ethnicity and school location were confounded in this study
because of the three data collection sites. Specifically, this was not a pure comparison of
the effects of location because of its relationship with race and education. On the other
hand, it accurately represented the separation that exists in mid—Michigan area schools.

This study was designed to examine the differences in the home-school
relationship and parent outcomes as a function of different school locations. However,
due to time and logistical constraints, data collection did not occur at a rural location.
Thus, only urban and suburban schools were compared. In addition, as noted previously,
the socioemotional and instrumental support comparisons were dropped because of high
multicollinearity.

However, this study still examined how social support influences trust,
satisfaction, and involvement for parents living in urban and suburban areas. Results
indicate that location does moderate the relationship between social support and parent
satisfaction. Again, an interaction term was first created by multiplying the dummy code
for school location (suburban schools comprised the reference group) by the social
support variable (Aiken & West, 1991). This term was entered following the social

support variable and the dummy code for school location. Again, values presented reflect

98

the true relationships between the variables, taking into account differences in direction
related to centering and logarithmically transforming variables.

Regression results indicated that the interaction term was significant for parent
satisfaction (standardized fl = .18, t = 2.12, p = .04). Because the interaction term was
significant, regression equations were plotted for parents living in suburban and urban
areas (Aiken & West, 1991). Low, mean, and high average values of social support were
plotted for both groups of parents. With higher levels of social support, suburban parents
experience more satisfaction than urban parents; with lower levels of support, suburban
parents experience less satisfaction than urban parents. Please see Table Nine for further
detail and please see Figure Three for a graphical representation of the interaction.

School location does not moderate the relationships between social support and
parent trust and parent involvement. The standardized regression coefficient for trust was
.01 (t = .11, p = .91) and the standardized regression coefficient for involvement was .17

(t = 1.38,p = .17). See Table Nine for further details.

99

Table 9
Summary of Hierarchical Regression Analysis for Social Support and School Location in

Predicting Parent Trust, Parent Satisfaction, and Parent Involvement

 

 

 

 

 

Parent Trust F Sig. AR2 B B T Sig.
Step 1 210.85 .00 .63

Social Support .48 .80 20.54 .00

Location .05 .03 .86 .39
Step 2 140.00 .00 .00

Social Support .49 .80 9.99 .00

Location .05 .03 .86 .39

Social Support x Location .01 .01 .11 .91
Parent Involvement F Sig. AR2 B B T Sig.
Step 1 19.65 .00 .14

Social Support .96 .15 2.50 .01

Location 5.22 .34 5.66 .00
Step 2 13.79 .00 .01

Social Support .01 .00 .02 .99

Location 5.29 .34 5.74 .00

Social Support x Location 1.25 .17 1.38 .17

 

100

Parent Satisfaction F Sig. AR2 B B T Sig.

 

 

 

Step 1 164.52 .00 .57
Social Support .40 .76 1 8.12 .00
Location .00 .00 .04 .97
Step 2 112.74 .00 .01
Social Support .48 .92 10.67 .00
Location .00 .00 .07 .95
Social Support x Location .11 .18 2.12 .04
Figure 3

Social Support and Parent Satisfaction as Moderated by School Location

Social Support and Parent Satisfaction as
Moderated by School Location

 

 

:A— Suburban. E
if ABLE

Parent Satisfaction

 

 

 

 

Low, Medium, and High Levels of
Social Support

101

In conclusion, this study’s results revealed that a number of findings. First, the PPSSS
and the global satisfaction measures are statistically valid. Further, this study suggests
that social support accounts for a sizeable amount of variance in parent trust and
satisfaction and a statically significant though small amount of variance in parent
involvement. The data also revealed that both parent education and school location
moderate the relationship between parent perceptions of school-provided social support

and parent satisfaction.

102

Chapter Five: Discussion

While research suggests that families are important in student achievement, the
literature supporting this connection is underdeveloped, unwieldy and conflicting. This
study sought to develop knowledge in this area and to untangle this complex relationship.

The current study yielded three major outcomes. First, this study offered some
evidence for social support as a psychological mechanism by which the home-school
relationship operates. However, the newly created measure of parent perceptions of
school-provided social support did not reveal a two or four part structure as suggested by
the psychological literature. In addition, this study indicated that parent perceptions of
the home-school relationship may consist of two factors rather than three as hypothesized
by Esler, Godber, and Christenson. Finally, this study shed light upon how the
relationship between social support and parent outcomes changed in the presence of
moderators including child grade, parent educational attainment, and school location.

Social Support as a Psychological Mechanism

As noted above, the first outcome of this study offered a novel explanation for the
relationship between families and schools. Namely, this study examined social support
as a psychological mechanism for the home—school relationship. This study tested the
theory that social support acts as a main effect, directly influencing parent outcomes.

Social support, a prominent psychological construct in the developmental and
clinical literatures, was conceptualized according to House’s (1981) theory to include
emotional, appraisal, instrumental, and informational support. In this study, social
support was considered as a single construct because of high intercorrelations between

the two major components, socioemotional support and instrumental support. These high

103

intercorrelations suggest that the four types of school-provided support are closely related
and that addressing the types of support separately is not appropriate in an elementary-
school parent population. Rather, parent perceptions of school-provided social support
should be viewed as a single construct that is theoretically but not empirically comprised
of four subtypes.

A number of reasons exist for why the content of social support consisted of one
factor rather than four as hypothesized by House (1981) or two as hypothesized by Thoits
(1982). First, this could be related to the level of measurement. Parents were asked to
rate their perceptions of the school as a whole including the teachers, principal, and staff.
When they consider the school as a whole, parents may not perceive receiving different
types of support. For instance, a teacher may provide emotional support while a principal
or school psychologist provides informational support. Another explanation is response
bias. It is likely that the most engaged and involved parents in each school responded to
the survey. This subset of the parent population tends to be involved regardless of
school-provided social support and may not perceive as many types of support because it
is not a determining factor in their involvement.

This single-component conceptualization of social support lends credence to
Sidney Cobb’s early model of social support. In “Social Support as a Moderator of Life
Stress,” Cobb defines social support solely in terms of information. The content of that
support is that the individual is loved, esteemed or valued, and that the individual
“belongs to a network of communication and mutual obligation” (Cobb, 1976, p. 300). In
short, Cobb hypothesizes that social support is one factor with a variety of components.

Although theoretically similar, Cobb’s research did not include Specific measures of

104

social support. Rather, he summarized numerous studies examining loose measures of
social support including wanted versus unwanted pregnancy and in-home versus hospital
care following surgery. In short, Cobb’s early single-factor conceptualization of social
support was re—affirmed in this study though the definitions varied dramatically.

This study suggests that social support might offer a reasonable explanatory
mechanism for the relationship between families and schools. Successful home-school
relationships rest upon two factors. First, schools provide support to parents by
conveying that they are respected and important and offering them assistance and
information. In addition, parents perceive the schools’ support. As perceived support
increases, parent trust and satisfaction increases.

This study’s finding that parent perceptions of social support are related to parent
outcomes lends credence to social support as a psychological mechanism underlying the
home-school relationship. In particular, these results suggest that LaRocco, House, and
French (1980) were correct in asserting that specific types of stress or strain, like parent
stress pertaining to education, are likely to be affected by a limited set of closely related
sources of support. In this case, parent trust and satisfaction, two measures that capture
parents’ affective relationship with the school, are directly related to the social support
provided by schools. In addition, this study confirms House’s assertion that informal,
high relevance forms of support are important and perhaps even more important than
formal sources of support. Although schools are being asked to play increasingly
complex roles in the lives of families, they were not created to support parents and are
thus informal sources of support. Schools are also of high relevance to parents because

of their frequent contact and importance in the lives of children.

105

Understanding the home-school relationship in terms of school-provided social
support has a number of implications for school practice and policy. As Lareau (1987)
noted, “When home-school relationships are evaluated exclusively in terms of parental
behavior, critical questions are neither asked nor answered” (p. 74). In keeping with this
observation, educators and researchers should seek to understand and influence the
schools’ role rather than focusing on parents. In particular, schools should seek to
provide parents with social support in all its forms.

In order for school personnel to provide socioemotional support, they should
examine their attitudes about parents. The literature on the sociology of education
suggests that three factors are essential. First, school personnel should assume that all
stakeholders have the child’s best interest in mind (Lewis & Forman, 2002). This basic
premise reminds parents and school staff that all parties are committed to supporting and
educating children. Second, school personnel should assume that all parents value
education and believe it is important for later success in life (Lott, 2001). Even though
parents may show their support of education in different ways, parents as a whole hold
this belief as fundamental. The third assumption school personnel should adopt is more
pragmatic: involvement at school is not a reflection of how much parents value education
(Lareau, 1987). Involvement is related to a variety of external and internal demands and
constraints (Lewis & Forrnan, 2002). Taken together, these attitudes and assumptions
make it possible for schools to provide parents with the socioemotional support they need
to feel like real partners in the home-school relationship.

Schools should also be cognizant of providing parents with instrumental support

in the form of information about education. Research on cultural capital, “the cultural

106

experiences in the home (that) facilitate children’s adjustment to school and academic
achievement” highlights parents’ need for information about education (Bordieu, 1977 as
cited by Lareau, 1987). Lareau noted that working-class parents tend to have less
information about schooling. This results from their social networks, which are
comprised of family members rather than other parents, and from the fact that their
children tend to be less involved in after-school activities, which results in fewer
opportunities for parents to become socialized to the culture of education (Lareau, 1987).

Schools can also improve instrumental support by asking parents what factors
prevent involvement and seeking to alleviate them. Lewis and Forrnan (2002) offer a
simple example of a school in a working class neighborhood offering free babysitting for
parent meetings at school. School meetings and phone conferences during non-business
hours also lessen some of the time constraints experienced by parents.

In short, social support has a direct influence on the relationship between parents
and schools. When schools conceptualize and structure their interactions with families to
provide socioemotional and instrumental support, families benefit and the home-school
relationship improves.

Examination of the Parent Perceptions of School-Provided Social Support Scafi

Along with a novel explanation for the home-school relationship, this study
included scale development as an outcome. Specifically, this study included the creation
of the Parent Perceptions of School-Provided Social Support (PPSSS) scale based on
Malecki and Demaray’s Child and Adolescent Social Support Scale (2000). The PPSSS
is a 24-item scale designed to capture parent perceptions of emotional, appraisal,

instrumental and informational social support. While empirical examination with a

107

diverse sample of parents suggests that this measure consists of one factor and is highly
reliable, the PPSSS failed to capture the four components of social support specified by
House (1981). All four parts were highly correlated and even a two-part
conceptualization, with socioemotional and instrumental support, did not yield separate
components statistically. As noted in the previous section, this could be related to scale
language; the PPSSS asked parents to consider the school as a whole, which eliminates
nuance in which members of the school community provide which types of support.
Also as noted above, it is likely that only the most active parents took the time to
complete and return the survey. Again, these parents are likely to be involved and
perceive support regardless of school efforts, which could eliminate the distinction in
perceived types of support.

This measure represents an important step forward in this literature because it
offers an initial attempt to capture school-provided social support. However, future
iterations of the measure should include substantive changes. In particular, the items
should be reworded to more explicitly capture the four types of support. Further, the
issue of source of support should be reconsidered (Tardy, 1985). Sources of support
should be separated into teachers, administrators, and other school professionals. As
research in this area progresses, participants should also include parents living in rural
areas as well as parents of children in middle through high school. Additionally, the
concurrent and predictive validity of this measure or future versions of this measure

should be further examined.

108

Examination of a Theoretical Model of Home-School Relationshfifi

Second, as noted above, this study suggests that Esler, Godber, and Christenson’s
2002 model may not best account for the structure of home-school relationships. In
particular, high multicollinearity between the attitudes and relationships components
suggest that they may not be separate factors. Instead, it may be more appropriate to
conceptualize parent perceptions of the home-school relationship in terms of affective
and behavioral components.

Data analyses revealed a significant, positive relationship between support and
satisfaction (attitudes) and trust (relationships). Specifically, social support accounted for
the most variance in trust, followed by satisfaction. This finding suggests that parent
perceptions of school-provided social support are important in parent trust and
satisfaction and vice versa. More specifically, schools that are intentional about
supporting parents have parents with more trust in their child’s teachers and parents who
experience more satisfaction with their child’s school. Further, the weak correlations
between parent involvement and social support, parent trust, and parent satisfaction might
indicate that actions are a separate factor in the home-school relationship at the
elementary level.

While attitudes and relationships have different practical significance, they were
not separate from a statistical standpoint in this parent population. Two possible
explanations exist for this finding. One explanation relates to the issue of the level of
analysis. The perceived social support measure and the parent satisfaction outcome

measure both asked parents to rate their perceptions of the school as a whole. At this

109

level of analysis, perhaps attitudes and relationships are indistinguishable to parents.
This leads to the second possible explanation for the gap between theory and data. The
data collected in this study only assessed parent perceptions of the home-school
relationship and parent outcomes of the relationship. This emphasis on parents leaves out
school personnel perceptions of the relationship, which may include both attitudes and
relationships. Given that Esler, Godber, and Christenson’s model was created from a
school perspective rather than a parent perspective, this gap makes sense.

Although the three-part model was not supported by data in this study, using the
Esler model as a framework still added richness to this study and advanced thinking
about the home-school relationship. This model called attention to the affective or
intangible component of the relationship between families and schools and reaffirmed a
nuanced approach to home-school collaboration. As noted in the review of the literature,
ample research exists detailing strategies to change school actions or strategies pertaining
to the home-school relationship (e.g., Baker, 2000). However, this atheoretical though
well-intentioned approach is not likely to be effective because it does not address the
processes underlying home-school relationships. In particular, schools can offer
numerous opportunities for parents to become involved but if parents do not see
themselves as critical for student learning or trust school staff, then the home-school
relationship will flounder. Esler, Godber, and Christenson’s theory serves as a reminder
that the affective parts of the relationship are as important, if not more so, than actions.

Moderators of the Home-School Relationship
The third major outcome of this study is that it showed how the relationship

between parent perceptions of support and parent satisfaction changed according to

110

parent education and school location. In particular, satisfaction with school varies more
for lesser-educated parents relative to how much social support they perceive. In
particular, when lesser-educated parents perceived low amounts of social support, they
reported lower satisfaction on average than did more-educated parents. When they
perceived greater social support, lesser-educated parents experienced more satisfaction on
average than did more-educated parents. This finding suggests that the satisfaction of
lesser-educated parents is more variable and that their satisfaction might be more
influenced by the efforts of schools to provide social support.

School location also acted as a moderator in this study. Parents whose children
attended suburban schools experienced lower mean levels of satisfaction than urban
parents when they perceived low levels of social support. Suburban parents perceived
higher mean levels of satisfaction than urban parents when they perceived high levels of
social support. This finding suggests that parent satisfaction with school might be more
open to school influences in suburban than urban parents. Again, suburban schools have
a unique opportunity to influence parent outcomes and their efforts to provide social
support are directly related to parent satisfaction.

Given the small amount of research on school location and parent education as
related to the home-school relationship, a new body of research was sought to make sense
of the findings. In particular, research on the sociology of education helps to explain the
interaction relationships between support and satisfaction for lesser-educated and
suburban parents (Lewis & Forman, 2002; Gorman, 1998; Lareau, 1987). This work
suggests that schools shape their interactions with parents and that school perceptions of

parents are vitally important (Lewis & Forman, 2002; Lott, 2001). First, a case will be

111

made for the overlap between working class parents in this literature and the lesser-
educated and urban parents in this study. Then, the intersection of middle class schools
and working class families will be presented. Finally, an example of how schools can
reframe their perceptions will be presented.

Literature on the sociology of education examines how working class parents
interact with the culture of education. These parents are typically those who live in urban
areas and those without advanced degrees. In particular, Lareau (1987) and others (e.g.,
Gorman, I998) label families as working class based on the communities where they
reside. Further, Lareau notes that working class parents tend to have a high school
diploma or not to have graduated high school. These descriptions suggest that the
working class label encompasses urban participants in this study and those with less
education.

Given the congruence between working class parents and parents in this study,
this literature can shed light upon the findings of this study. In particular, research on the
sociology of education suggests that there is a disconnection between schools and
working class parents, but that it can be ameliorated through how schools perceive and
work with families.

First, schools are middle class institutions. As such, middle class beliefs color
their relationships with working class parents (Lareau, 1987). One school-held belief is
that working class parents do not foster educational achievement (Lott, 2001). While this
stereotype is true in some regards, Gorrnan’s 1998 research indicated that about half of

the working class parents he interviewed held middle class ideas about the value of

112

education and attaining a college degree. Those parents who do not conform to these
beliefs will require more targeted efforts on the part of schools.

On the other hand, when schools change their perceptions of parents and
involvement, this relationship can be highly successful. One example of schools tapping
into parent strengths rather than focusing on unmet expectations is redefining
involvement in terms of quality versus quantity. While schools often focus on the
amount of time parents spend in schools, this is a tiny fraction of involvement (Grolnick
& Slowiaczek, 1994). Schools are correct in noting that working class parents respond
less to teacher requests for participation in elementary schools (Lareau, 1987; Lewis &
Forman, 2002). This decreased quantity is related to a variety of factors like rigid work
schedules, significant family demands, and the necessity of having multiple wage earners
per family. Despite these constraints, the quality of involvement is similar for working
and middle class parents (Lewis & Forman, 2002). When they volunteer at school,
parents from the working class engage in the same types of activities with similar
amounts of success (Lewis & Forman, 2002). In fact, Lewis and Forrnan (2002) found
that far more struggles erupted at an upper class school than at a working class school
over issues including “resources, autonomy, respect, and authority” (p. 73).

This example suggests that how schools conceptualize the home-school
relationship and their expectations of parents might play a role in parent satisfaction with
school. When their efforts are noted and appreciated, parents are likely to be more
satisfied. Further research should be done on how parent perceptions of how their
involvement is received contribute to satisfaction. Schools too should make every effort

to recognize the class differences and to seek to reduce them through valuing parent

113

efforts and by reaching out to parents, and particularly those who belong to the working
class. Specific strategies are detailed below.

Implications of the Study as a Whole: Affective and Behavioral Strategies for Improving
the Home-School Relationship

As noted above, literature on the sociology of education suggests that the
relationship between parents of different classes and schools is complex. Research also
indicates that schools strongly influence the parents of their students (Goldring, 1990;
Ball, Bowe, & Gewirtz, 1995; Lott, 2001; Lewis & Forman, 2002). Given the influence
schools have on parents, schools offer a promising route to improving the home-school
relationship. One way to organize these intervention strategies is through the affective
and behavioral framework suggested by the parent outcome data.

Affect comprises the heart of the home-school relationship (Esler et al., 2002). In
order for schools to engage more effectively with families, they should be intentional
about the affective strategies they use. Two examples of affective strategies include a
strengths-based perspective and expanding the definition of family.

The first step schools should take to improve the affective side of their
relationship with families is to highlight student and parent strengths. Much
communication between schools and parents focuses on student behavioral challenges
and academic weaknesses (Jones & Jones, 2001). In particular, schools are most likely to
contact parents when their children are not meeting expectations (Lott, 2001). This can
sour the relationship between families and schools by building up negative associations.
When schools communicate student strengths and growth, parents feel encouraged and

successful. Schools can also note when parents have been particularly effective in

114

contributing to a child’s learning. In addition to helping parents feel valued, this can help
schools to shift their focus from deficits and remediation to strengths and success.
Schools can also improve the home-school relationship by expanding their
definition of family. All important adults in a child’s life should be welcome at school
open houses, student conferences, and educational meetings. This policy turns working
class family networks into a valuable educational asset. It also relieves parents of some
of the burdens of time constraints and rigid schedules and acknowledges that not all
parents are available during regular business hours. This policy also allows school
personnel to form relationships with a larger proportion of the influential people in a
child’s life. As one caveat, parents should be responsible for designating which adults
can attend school events and meetings so that the safety of students and staff is ensured.
In addition to targeting the affective side of the relationship, schools can change
their behaviors to support families. In particular, two actions might be effective in
improving the home-school relationship. First, schools can assist parents in forming
organizations related to educational issues (Goldring, 1990). As Goldring notes, “parents
are more likely to be influential in the school organization if they are formally organized
as interest groups” (1990, p. 5). Parent outreach and cuniculum development are two
examples of where parents could be influential. These organizations allow parents to
develop real power in the school and to feel as if they are more than stakeholders. Parent
participation in these organizations also allows parents to see that they are necessary for
successful education and somewhat reduces the power differential between school

personnel and parents. Finally, forming groups helps parents to create networks with

115

other parents, which helps them to navigate the culture of school and to feel more integral
to its functioning (Lewis & Forman, 2002).

A second action schools can take to improve family-school relationships is to
become community social service centers (Lott, 2001; Zigler & Gilman, 1991). Zigler
and Gilman conceptualize these twenty-first century schools as educational, childcare,
and family support centers. In particular, Zigler notes that numerous services exist to
support parents and families and that these services are highly fragmented. Further,
Zigler and Gilman assert that having schools coordinate services would prevent
duplication of services, would ensure that families had access to relevant services, and
would help to improve relationships between families and schools. This model has
already been adopted in some areas with some success (Lott, 2001; Zigler & Gilman,
1991). While this action requires policy shifts and substantial investment of resources, it
has the potential to improve parent perceptions of the schools and service delivery to
parents and children. First, school-based community centers could make schools seem
more caring (Lott,2001). By meeting the basic needs of families, these centers convey to
parents that they are important beyond their role as home caregiver. In addition, these
centers would help schools to expand their view of children beyond mere learners. In
other words, these centers could help schools to better understand the risk and protective
factors in their students’ lives (Masten & Coatsworth, 1998). Finally, these centers
would help families to receive seamless coverage without service duplication or gaps.

In short, school strategies are vital in determining the success of the home-school
relationship. Through affective strategies such as highlighting strengths, welcoming

important adults, and bringing families into schools as well as behavioral strategies such

116

as helping parents to form organizations and becoming community service centers,
schools can engage parents and contribute to their satisfaction.
Limitations and Additional Research

Some limitations exist pertaining to the selected schools, the participants, and the
methods in this study. The first challenge was that the three schools in this study were
clearly divided into two categories: suburban and urban. Within these schools, parent
race and education were confounded. Hence, the race analysis was dropped from the
study, which makes it harder to determine whether the close relationship between school
location and parent education are widespread or are related to the particular schools in
this study. Future research including more heterogeneous parent populations might
clarify these results.

A second limitation pertained to the parent participants. Parents who are willing
to read, complete, and return a five-page survey are likely the most involved and
organized parents in the school. Since no more than thirty-one percent of the parents at
any of the schools responded, most parent perceptions were not represented. The results
of this study would likely differ if the other seventy percent of the parents had responded.
This was also evident in the fact that there was a positive response bias in the survey
responses with many parents selected the highest mark for perceived school-provided
social support, trust, and satisfaction. More parent perspectives might have been
included had the researcher attended parent-teacher conferences or followed up survey
distribution with phone calls. However, the researcher wished to respect parent privacy
.while conducting this research. Attending conferences and adding parent phone calls

should be included in future data collection.

117

A third area of possible limitations pertained to the study method. The survey
was structured so that parents would consider the school as an organization. While this is
consistent with the literature, grouping the “teachers, principal, and staff” obscured
individual contributions to parents’ experiences. This was highlighted in some of the
comments parents wrote in the margins of their surveys. In the parent trust questions, a
parent marked low trust scores and wrote, “This information is about my child’s teacher
this year. Had I filled the survey out the last two years, I would have marked a lot more
45 and Ss.” Another parent called to suggest that I separate questions pertaining to the
principal, the current teacher, and the school staff. She noted that her responses would
differ dramatically based on whom she was rating.

A final limitation is that this study only captured parent perceptions at one point
in time. This snapshot did not allow for the detection of causal relationships between the
constructs. Thus, it was impossible to determine whether parent perceptions of school-
provided social support actually caused increased trust and satisfaction and whether the
moderation relationships were directly related to school efforts to enhance the home-
school relationship.

Future Directions

While this study shed light on the home-school relationship, a number of new
questions were raised in the course of the research and in reflection on the limitations.
These questions and limitations point to both conceptual and methodological directions
for future research.

The first conceptual challenge related to the theoretical framework for this study.

In particular, this study indicated that the attitudes, relationships, and actions framework

118

proposed by Esler and colleagues (2002) was not an effective basis for the home-school
relationship. In particular, only an affective and a behavioral component seemed to exist.
Future measures of the home-school relationship should take into account these two
factors and it should be used to guide scale development.

A second conceptual issue is that only parent perceptions of the home-school
relationship were assessed in this study. Given that Esler, Godber, and Christenson
(2002) write from a within—school perspective, one that stresses the interaction between
families and schools, this is a glaring omission. In future studies, the perspectives of
school personnel, and particularly teachers and administrators, should be assessed. This
would help to determine how school perceptions fit with parent perceptions and where
discrepancies lie. School perspectives on the outcomes of the relationship should also be
measured. Again, given that schools and families are the two largest stakeholders in this
relationship, it should be captured from multiple perspectives.

In addition, three methodological issues could be addressed in future research.
First, future research in this area should expand beyond a quantitative framework.
Converting the Home-School Relationship Survey into a structured interview or an open-
ended questionnaire would allow parents more flexibility and would more accurately
capture parents’ voices. First, parents could clearly specify about whom they are
responding. In addition, more flexible questions about their perceptions of support might
more clearly delineate what types of support parents value and if this changes with
demographic factors. Parent responses could also add richness to the theory supporting
parent perceptions of school-provided social support. One drawback with an open-ended

format is that parents might be less inclined to complete the questionnaires since it would

119

require more effort and increased literacy. Also, interviews require a larger time
commitment than Likert—type questions and can be more intrusive.

Second, future efforts should examine support provided by different members of
the school community. The PPSSS scale used in this study asked parents to rate support
from their “child’s school (including the teachers, principal, and staff)” This was
intentionally worded in general, inclusive terms so parents could define school in their
own terms. As noted above though, parent participants indicated that members of the
school community offered different types and levels of support.

One challenge is that separating out different members of the school community
might cause parents to reflect on individuals rather than the school as an institution. This
could cause parent perceptions of the school and its staff to vary dramatically from yeaf
to year and to be highly affected by the personalities of involved individuals.
Information about the school as a whole needs to be balanced with information about
how different school personnel in particular roles contribute to their relationship with
families. Parents might expect different types of support from different staff members
and more specificity about parent expectations and perceptions might help to tailor more
specific interventions.

Finally, collecting data at two points in time would allow for the detection of
changes in the home-school relationship and for the determination of whether school
efforts influenced parent perceptions of support or parent outcomes. It would also
indicate whether school interventions caused changes in parent perceptions of social

support and if an increased sense of support contributed to changes in parent outcomes.

120

Overall, this study suggests that school-initiated actions influence the satisfaction
of parents with school. Thus, schools should make every effort to reach out to parents
and to support them in ways that maximize their cultural capital. These supportive
attitudes, relationships, and actions have the potential to influence parent satisfaction.
Further, schools should target lesser-educated and suburban parents in their efforts since
their satisfaction varies more with their perceptions of support. In short, schools are
responsible for supporting parents such that the home-school relationship is maintained

and student learning is enhanced.

121

Appendix One: Teacher Letter
Dear Teachers,

You are invited to participate in a research study with the School Psychology
Program at MSU.

What is the purpose (3 the research?

This research is designed to help us understand how the school's relationship
with parents affects parent satisfaction with school, trust of the school,
and involvement in school activities.

WhaLt woridfiyog like me to do?
0 Send home the surveys with your students
. Remind students to have their parents complete the surveys
- Send home duplicate surveys when necessary

Thank you so much for your participation! Please email or call Jana
Aupperlee (AUPPERLBC‘Pmsgedju or 485-4865) or Jean Baker
(ibakeersuedu or 432-0843) if you have any questions about your
participation or the study.

Sincerely,
Jana Aupperlee

122

Appendix Two: Parent Letter

Attention Parents
You are invited to participate in a research study with the School
Psychology Program at MSU! We will be working together for the
rest of the school year to examine how the relationship between
families and schools influences parents.

What is the purpose of the research? This research is designed
to help us understand how support from the school affects
parent satisfaction with school, trust of the school, and
involvement in school activities.

What will I do? In a few days, your child will bring home a packet
from school. You will be asked to complete a short survey about
your relationship with your child's school. Then, you will return
the survey to the researchers in the provided envelope.

What about privacy? No one will see your answers to the
questions except the researchers. Also, your personal
information and responses to the questions will be separated and
destroyed when the survey is completed. You and your child will
not be identified in any way.

How can I be included? You can only be included if you sign and
return the permission slip with the survey in the provided
envelope.

Thank you! Please e-mail or call Jana Aupperlee
(AUPPERL3@msu.edu or 485-4865) or Jean Baker
(ibaker@msu.edu or 432-0843) if you have any questions about
your participation or the study.

Please mail the survey and peach permission slip by April 7, 2005.

123

Appendix Three: Home-School Relationship Survey

Dear Parent,
Please complete the peach-colored consent form and the following survey and return
them both in the postage-paid envelope. As you complete this survey, please think about one of

your children. If you have two or more children in this school, please think about your
YOUNGEST child in kindergarten through fifth grade.

Thank You!
Jan u I .erlee

 

Home-School Relationship Survey
0 In what grade is your youngest child in this building? Kindergarten I" 2"d 3"d 4th 5th
0 What is your gender? Male Female
0 What is the gender of the child you’re thinking about for this survey? Male Female
0 Does your child receive special education services? Yes No

0 What is your race/ethnicity?
o Caucasian

0 African American
0 Hispanic

0 Asian

0 Other:

 

0 Please circle your highest education level:
0 Some high school
High school graduate or GED
Some college
Associates degree
Bachelor’s degree
Some post-college courses
Graduate degree (including: M.A., M.D., Pb.D., etc.)

000000

124

 

  

 

 

 

My child’s school (including the teachers, principal, and staff)...

cares about me (Not at All) 1 2 3 4
understands me (Not at All) 1 2 3 4
listens when I need to talk (Not at All) 1 2 3 4
treats me with respect (Not at All) 1 2 3 4
understands my child (Not at All) 1 2 3 4

respects my contributions to my child’s learning
(NotatAll)1 2 3 4

answers my questions (Not at All)l 2 3 4

gives me helpful advice about my child’s learning and development
(Not at All) 1 2 3 4

explains thingsldon’t understand (Not at All)l 2 3 4

helps me solve problems by giving me information
(Not at All) 1 2 3 4

shows me how to do things (Not at All) 1 2 3 4

gives me information about programs and services
(Not at All)l 2 3 4

appreciates my help at school and in the classroom
(Not at All) 1 2 3 4

lets me know I’m doing a good job as a parent
(Not at All) 1 2 3 4

has clear expectations for my participation in my child’s learning
(Not at All) 1 2 3 4

thinks parents are important for learning (Not at All) 1 2 3 4

appreciates my work with my child at home
(Not at ND] 2 3 4

125

Please rate your response to the following statements about your child’s school:

Iagree... 1(Not at All) 2(A Little) 3(Somewhat) 4(Mostly) 5( Definitely)

  

5 (Definitely)
5 (Definitely)
5 (Definitely)
5 (Definitely)

5 (Definitely)

5 (Defmitely)

5 (Definitely)

5 (Definitely)

5 (Definitely)

5 (Definitely)

5 (Definitely)

5 (Definitely)

5 (Definitely)

5 (Definitely)

5 (Definitely)

5 (Definitely)

5 (Definitely)

 

My child’s school (including the teachers, principal, and staff)...

0 gives me feedback about my child’s learning, behavior, and social interactions
(Not at All) 1 2 3 4 5 (Definitely)

0 spends time with me whenlneed help (Not at All)l 2 3 4 5(Definitely)

0 provides programs to make sure my child learns
(Not at All) 1 2 3 4 5 (Definitely)

0 makes sure that my child has the things they need to learn (like books or school supplies)
(Not at All) 1 2 3 4 5 (Definitely)

0 helps me make decisions about my child’s education
(Not at All) 1 2 3 4 5 (Definitely)

0 provides programs to help parents do a better job
(Not at All) 1 2 3 4 5 (Definitely)

0 helps me find resources (like information about after-school programs or summer camps
(Not at A10] 2 3 4 5 (Definitely)

Please rate your agreement with the following statements about your child’s teacher(s):

 

Iagree... 1(Not at All) 2(A Little) 3(Somewhat) 4(Mostly) 5(Definitely)

I trust that teachers...

are doing a good job teaching my child academic subjects
(Not at All) 1 2 3 4 5 (Definitely)

are doing a good job teaching my child to follow rules and directions
(Not at All) 1 2 3 4 5 (Definitely)

are doing a good job helping my child resolve conflicts with peers
(Not at All) 1 2 3 4 5 (Definitely)

are doing a good job keeping me well informed of my child’s progress
(Not at All) 1 2 3 4 5 (Definitely)

are doing a good job encouraging my participation in my child’s education
(Not at All) 1 2 3 4 5 (Definitely)

are doing a good job disciplining my child (Not at All) 1 2 3 4 5 (Definitely)

are doing a good job encouraging my child’s sense of self-esteem
(Not at All) 1 2 3 4 5 (Definitely)

are doing a good job encouraging my child to have a positive attitude toward learning
(Not at All) 1 2 3 4 5 (Definitely)

126

I trust that teachers...

0 are doing a good job helping my child understand his/her moral and ethical responsibilities
(Not at All) 1 2 3 4 5 (Definitely)

are friendly and approachable (Not at All)l 2 3 4 5(Definitely)

are receptive to my input and suggestions (Not at All)l 2 3 4 5(Definitely)

0 are sensitive to cultural differences (Not at All) 1 2 3 4 S (Definitely)
0 respect me as a competent parent (Not at All) 1 2 3 4 5 (Definitely)
0 care about my child (Not at All) 1 2 3 4 5 (Definitely)
0 have my child’s best interests at heart (Not at All) 1 2 3 4 5 (Definitely)
0 are worthy of my respect (Not at All) 1 2 3 4 5 (Definitely)

0 will do what is best for my child in the classroom
(Not at All) 1 2 3 4 5 (Definitely)

0 are easy to reach when I have a problem or question
(Not at All) 1 2 3 4 5 (Definitely)

0 keep me aware of all the information I need related to school
(Not at All) 1 2 3 4 5 (Definitely)

Please rate your agreement with the following statements about your child’s school:

 

Iagree... 1(Not at All) 2(A Little) 3(Somewhat) 4(Mostly) 5(Definitely)

o This school is a good place for my child (Not at All) 1 2 3 4 5 (Definitely)

0 The school staff is doing good things for my child
(Not at All) 1 2 3 4 5 (Definitely)

0 l have confidence in the people at this school
(Not at All) 1 2 3 4 5 (Definitely)

0 This school is preparing my child for the future
(Not at All) 1 2 3 4 5 (Definitely)

127

Please rate your agreement with the following statements about your child’s school:

 

1(Not Satisfied) 2(Somewhat Satisfied) 3(Satisfied) 4(Mostly Satisfied) 5(Very Satisfied)

How satisfied are you with...
0 the school’s beliefs and attitudes about learning?

(Not Satisfied)1 2 3 4 5(Very Satisfied)
0 the school’s friendliness? (Not Satisfied)l 2 3 4 5(Very Satisfied)

0 the school’s promotion of your child’s learning and academic development?
(Not Satisfied) l 2 3 4 5 (Very Satisfied)

o the school’s promotion of your child’s ability to get along with others?
(Not Satisfied) 1 2 3 4 5 (Very Satisfied)

0 the school’s promotion of your child’s knowledge of how to be a good citizen?
(Not Satisfied) l 2 3 4 5 (Very Satisfied)

0 the school’s development of your child’s learning potential?
(Not Satisfied) 1 2 3 4 5 (Very Satisfied)

0 the school’s curriculum or what it is teaching your child?
(Not Satisfied) l 2 3 4 5 (Very Satisfied)

o the school’s promotion of your values and beliefs?
(Not Satisfied) l 2 3 4 5 (Very Satisfied)

Please rate how oflen you participate in the following activities:

 

How often do you...
0 read to your child? (Never) 1 2 3 4 5 (All the Time)

0 take your child to the library? (Never) 1 2 3 4 5 (All the Time)
0 play educational games with your child? (Never) 1 2 3 4 5 (All the Time)
0 visit school for special events? (Never) 1 2 3 4 5 (All the Time)

0 attend PTO (Parent-Teacher Organization) meetings?

(Never) 1 2 3 4 5 (All the Time)
0 send things to class for your child (books, snacks, etc.)?
(Never) 1 2 3 4 5 (All the Time)

How often do you...

0 volunteer at your child’s school? (Never) 1 2 3 4 5 (All the Time)
0 call your child’s teacher? (Never) 1 2 3 4 5 (All the Time)
0 write notes or email to your child’s teacher? (Never) 1 2 3 4 5 (All the Time)
0 stop to talk to your child’s teacher? (Never) 1 2 3 4 5 (All the Time)
0 attend parent-teacher conferences? (Never) 1 2 3 4 5 (All the Time)

129

Appendix Four: Parent Consent

Parent’s Name: Child’s Name: Grade:

PARENT/GUARDIAN CONSENT FORM

As a parent at insert school name here, you are invited to participate in this study called “Home-
School Relationships.” This study examines how the relationship between families and schools
influences parent trust of school, satisfaction with school, and involvement.

1) The reason for this research is to understand how the relationship between the families and
schools contributes to parent outcomes. From this research, we hope to understand what types of
support are important to parents and how schools can better support families.

2) If you choose to complete this study, you will be asked to complete a short (15 minute) survey
about your perceptions of your child’s school.

3) Your participation is completely voluntary. You may choose not to participate at all, you may
refuse to answer certain questions, or you may discontinue your participation at any time without
penalty. No discomforts or stresses are foreseen.

4) We do not anticipate that you will be stressed or uncomfortable as a result of their participation
in this research. However, if you do so, we are prepared to refer you to a knowledgeable
professional.

5) The results of your participation will be confidential, and will not be released in any
individually identifiable from without your prior consent, unless otherwise required by law. No
one will be able to identify your results from this study. Refusal to participate or withdrawal
from participation will not in any way penalize you. You may have the results of the
participation, to the extent that they can be identified as yours, returned to you, removed from the
research records, or destroyed at any point prior to the end of the study.

130

The investigators will answer any further questions about the research, now or during the course
of the project. You are encouraged to ask questions. You may talk with anyone on the research
team during the study, or you may contact the researchers:
- Jana Aupperlee, B.A.; Michigan State University, Department of Counseling Psychology,
Educational Psychology, and Special Education, East Lansing, MI, 48824.
Telephone: (517) 485-4865. E—mail: aupper13@msu.edu

0 Dr. Jean A. Baker; Michigan State University, Department of Counseling Psychology,
Educational Psychology, and Special Education, East Lansing, MI, 48824.
Telephone: (517) 432-0843. E-mail: jbaker@msu.edu

If you agree to participate in the research, please sign below and return this form, with your
questionnaire, in the attached postage-paid envelope.

 

 

Signature of Parent/Guardian Date Signatures of Investigators Date

 

 

Research at Michigan State University involving human participants is overseen by the
University Committee on Research Involving Human Subjects. If you have any questions or
concerns regarding your rights as a participant, or are dissatisfied at any time with any aspect of
the study, you may contact —- anonymously, if you wish — Peter Vasilenko, Pb.D., Chair of the
University Committee on Research Involving Human Subjects by phone: (517) 355-2180, fax:
(517) 432-4503, e-mail: ucrihs@msu.edu. or regular mail: 202 Olds Hall, East Lansing, MI
48824.

 

131

 

Appendix Five: Reminder Letter

Dear Parents,

Last week, you received a manila envelope containing the
Home-School Relationship Survey. Have you signed the
peach-colored consent form and returned the survey
yet? If not, please do so today!

Remember, this research will help us understand how
support from the schools influences parent feelings
about their child's school and their involvement with
school activities!

If you have any questions about your participation or the
study, please call or send us an e—mail:
0 Jana Aupperlee:
(517)485-4865 or AUPPERL3@msu.edu
- Jean Baker
(517) 432—0843 or J BA KER@msu.edu

Thank You,
Jana Aupperlee

132

References

Adams, K. S. & Christenson, S. L. (2000). Trust and the family-school
relationship examination of parent-teacher differences in elementary and secondary
grades. Journal of School Psychology, 38, 477-497.

Aiken, L. S. & West, S. G. (1991). Multiple regression: Testing and interpreting
interactions. London: Sage.

Ames, C. (1993). How home-to-school communication influences parent beliefs
and perceptions. Equity and Choice, 9, 44-49.

Anderson, E. S. & Keith, T. Z. (1997). A longitudinal test of a model of
academic success for at-risk high school students. Journal of Educational Research, 90,
259-268.

Auerbach, S. M. & Kilmann, P. R. (1977). Crisis intervention: A Review of
outcome research. Psychological Bulletin, 84, 1189-1217.

Babbie, E. (1990). Survey Research Methods. Belmont, CA: Wadsworth.

Baker, A. J. (2000). Making the promise of parent involvement a reality. High
School Magazine, 7, 14-17.

Ball, S. J ., Bowe, R., & Gewirtz, S. (1995). Circuits of schooling: A sociological
exploration of parental choice of school in social class contexts. The Sociological
Review, 43, 52-78.

Baron, R. M. & Kenny, D. A. (1986). The moderator — mediator variable
distinction in social psychological research: Conceptual, strategic, and statistical

considerations. Journal of Personality and Social Psychology, 51, 1173-1182.

Bempechat, J. (1992). The role of parent involvement in children’s academic
achievement. School Community Journal, 2, 31-41.

Boekaerts, M. (1999). Self-regulated learning. International Journal of
Educational Research, 31, 445-551.

Bowlby, J. (1969). Attachment and Loss. New York: Basic Books.

Bowlby, J. (1980). Attachment and Loss with Additional Notesfrom the Author.
New York: Basic Books.

Bronfenbrenner, U. (1977). Toward an experimental ecology of human
development. American Psychologist, 32, 513-531.

133

Caplan, G. (1974). Support systems and community mental health: Lectures on
concept development. New York: Behavioral Publications.

Cauce, A. M., Felner, R. D., & Primavera, J. (1982). Social support in high-risk
adolescents: Structural components and adaptive impact. American Journal of
Community Psychology, 10, 417-428.

Cauce, A. M. & Srebnik, D. S. (1990). Returning to social support systems: A
Morphological analysis of social networks. American Journal of Community Psychology,
18, 609-616.

Cassel, J. (1976). The Contribution of the social environment to host resistance:
the Fourth Wade Hampton Frost lecture. American Journal of Epidemiology, 104, 107-
123.

Christenson, S. L. & Buerkle, K. (1999). Families as educational partners for
children’s school success: Suggestions for school psychologists. In C. R. Reynolds & T.
B. Gutkin (Eds), The Handbook ofschool psychology (3“ ed.) (pp. 709-744). Bethesda,
MD: National Association of School Psychologists.

Christenson, S. L. & Godber, Y. (2001). Enhancing constructive family-school
connections. In J. N. Hughes & A. M. La Greca (Eds), Handbook of psychological
services for children and adolescents (pp. 455-476). New York: Oxford University
Press.

Cobb, S. (1976). Social support as a moderator of life stress. Psychosomatic
Medicine, 38, 300-314.

Cohen, S. & Wills, T. A. (1985). Stress, social support, and the buffering
hypothesis. Psychological Bulletin, 98, 310-357.

Creswell, J. W. (2003). Research design: Qualitative, quantitative, and mixed
methods approaches. Thousand Oaks, CA: Sage.

Cutrona, C. E. & Russell, D. W. (1990). Type of social support and specific
stress: Toward a theory of optimal matching. In R. B. Sarason, I. G. Sarason, & G. R.
Pierce (Eds), Social support: an Interactional view (pp. 319-366). Oxford, England:
Wiley and Sons.

Demaray, M. K. & Malecki, C. (2002). Critical levels of perceived social support
associated with student adjustment. School Psychology Quarterly, 1 7, 213-241.

Desimone, L. (1999). Linking parent involvement with student achievement: Do
race and income matter? Journal of Educational Research, 93, 11-30.

134

Dubow, E. F., Tisak, J ., Causey, D., Hryshko, A. & Reid, G. (1991). A Two-year
longitudinal study of stressful life events, social support, and social problem-solving
skills: Contributions to children’s behavioral and academic adjustment. Child
Development, 62, 583-599.

Eccles, J. S. & Harold, R. D. (1993). Parent-school involvement during the early
adolescent years. Teachers College Record, 94, 568-587.

Epstein, J. (1995). School/family/community partnerships: Caring for the children
we share. Phi Delta Kappan, 76, 701-712.

Epstein, J. L. & Dauber, S. L. (1991). School programs and teacher practices of
parent involvement in inner-city elementary and middle schools. Elementary School
Journal, 91, 289-305.

Esler, A. N., Godber, Y., & Christenson, S. (2002). Best practices in supporting
home-school collaboration. In A. Thomas & J. Grimes (Eds), Best practices in school
psychology IV (pp. 389-411). Bethesda, MD: National Association of School
Psychologists.

Fan, X. & Chen, M. (2001). Parental involvement and students academic
achievement: A Meta-analysis. Educational Psychology Review, 13, 1-22.

F ehrrnann, P. 6., Keith, T. Z., & Reimers, T. M. (1987). Home influence on
school learning: Direct and indirect effects parental involvement on high school grades.
Journal of Educational Research, 80, 330-337.

Folkman, S. & Lazarus, RS. (1980). An analysis of coping in a middle-aged
community sample. Journal of Health and Social Behavior, 21, 219-239.

F euerstein, A. (2000). School characteristics and parent involvement: Influences
on participation in children’s schools. Journal of Educational Research, 94, 29-39.

Goldring, E. B. (1990). Principals’ relationships with parents: The homogeneity
versus the social class of the parent clientele. The Urban Review, 22, 1-15.

Goldring, E. & Shapira, R. (1993). Choice, empowerment, and involvement:
What satisfies parents? Educational Evaluation and Policy Analysis, 15, 396-409.

Gorman, T. J. (1998). Social class and parental attitudes toward education:
Resistance and conformity to schooling in the family. Journal of Contemporary
Ethnography, 27, 10-44.

Gottfi'ied, A. E., Fleming, J. S., Gottfried, A. W. (1994). Continuity of academic

intrinsic motivation from childhood through late adolescence: A Longitudinal study.
Journal of Educational Psychology, 86, 104-113.

135

Griffith, J. (1997). School and parent perceptions of school social environment:
Are they group based? The Elementary School Journal, 98, 135-150.

Griffith, J. (1998). The relation of school structure and social environment to
parent involvement in elementary schools. The Elementary School Journal, 99, 53-80.

Griffith, J. (2000). School climate as group evaluation and group consensus:
Student and parent perceptions of the elementary school environment. The Elementary
School Journal, 101, 35-61.

Grolnick, W. S. & Slowiaczek, M. L. (1994). Parents’ involvement in children's
schooling: A multidimensional conceptualization and motivational model. Child
Development, 65, 237-252.

Hara, S. R. & Burke, D. J. (1998). Parent involvement: The key to improved
student achievement. School Community Journal, 8, 9-19.

Hausman, C. & Goldring, E. (2000). Parent involvement, influence, and
satisfaction with magnet schools: Do reasons for choice matter? The Urban Review, 32,
105-121.

Hayes, S. C., Barlow, D. H., & Nelson-Gray, R. O. (1999). The Scientist
Practitioner: Research and Accountability in the Age of Managed Care. Boston: Allyn
and Bacon.

Henderson, A. & Berla, N. (Eds). (1997). A new generation of evidence: The
family is critical to student achievement. Washington, DC: Center for Law and
Education.

Holmbeck, G. (1997). Toward terminological, conceptual, and statistical clarity
in the study of mediators and moderators: Examples from the child-clinical and pediatric
psychology literatures. Journal of Consulting and Clinical Psychology, 63, 599-610.

Hoover-Dempsey, K. V., Bassler, O. C., & Brissie, J. S. (1987). Parent
involvement: Contributions of teacher efficacy, school socioeconomic status, and other
school characteristics. American Educational Research Journal, 24, 417-435.

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

Howell, D. c. (2002). Statistical Methods for Psychology (5” Edition). Pacific
Grove, CA: Duxbury.

136

1220, C. V., Weissberg, R. P., Kasprow, W. J., & Fendrich, M. (1999). A
longitudinal assessment of teacher perceptions of parent involvement in children’s
education and school performance. American Journal of Community Psychology, 27,

817-839.

Jones, V. F. & Jones, L. S. (2001). Comprehensive classroom management:
Creating communities of support and solving problems. Boston: Allyn and Bacon.

Keith, T. Z., Keith, P. B., Quirk, K. J., Sperduto, J ., Santillo, S., & Killings, S.
(1998). Longitudinal effects of parent involvement on high school grades: Similarities
and differences across gender and ethnic groups. Journal of School Psychology, 36, 335-
363.

Kohl, G. O., Lengua, L. J., & McMahon, R. (2000). Parent involvement in school:
Conceptualizing multiple dimensions and their relations with family and demographic
risk factors. Journal of School Psychology, 38, 501-523.

LaRocco, J. M., House, J. S., & French, J. R. P. (1980). Social support,
occupational stress, and health. Journal of Health and Social Behavior, 21, 202-218.

Lazarus, R. S. (1966). Psychological stress and the coping process. New York:
McGraw-Hill.

Lewis, A. E. & Forman, T. A. (2002). Contestation or collaboration? A
comparative study of home-school relations. Anthropology and Education Quarterly, 33,
60-89.

Lott, B. (2001). Low-income parents and the public schools. The Journal of
Social Issues, 5 7, 247-259.

Luthar, S. S., Cicchetti, D., & Becker, B. (2000). The Construct of resilience: A
Critical evaluation and guidelines for future work. Child Development, 71, 543-562.

Malecki, C. K. & Demaray, M. K. (2002). Measuring perceived social support:
Development of the Child and Adolescent Social Support Scale. Psychology in the
Schools, 39, 1-18.

Marcon, R. (1999). Positive relationships between parent school involvement and
public school inner-city preschoolers’ development and academic performance. School
Psychology Review, 28, 395-412.

Mash, E. J. & Dozois, D. J. A. (2003). Introduction. In E. J. Mash and R. A.
Barkley (Eds), Child Psychopathology (3-74). New York: Guilford.

137

Masten, A. M. (1994). Resilience in individual development: Successful
adaptation despite risk and adversity. In M. Wang and E. Gordon (Eds), Risk and
resilience in inner city America: Challenges and prospects (pp.3-25). Hillsdale, NJ:
Erlbaum.

Masten, A. M. & Coatsworth, J. D. (1998). The development of competence in
favorable and unfavorable environments. American Psychologist, 53, 205-220.

Masten, A. 8., Best, K. M., & Garmezy, N. (1990). Resilience and development:
Contributions from the study of children who overcome adversity. Development and
Psychopathology, 2, 425-444.

Masten, A. S., Hubbard, J. J ., Gest, S. D., Tellegen, A., Garmezy, N., & Ramirez,
M. (1999). Competence in the context of adversity: Pathways to resilience and

maladaptation from childhood to late adolescence. DeveIOpment and Psychopathology,
I I, 143-169.

McNeal, R. B. (1999). Parent involvement as social capital: Differential
effectiveness on science achievement, truancy, and dropping out. Social Forces, 78, 117-
144.

Pallant,J. (2001). SPSS Survival Manual. Buckingham, England: Open
University Press.

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, 908-923.

Rempel, J. K., Holmes, J. G. & Zanna, M. P. (1985). Trust in close relationships.
Journal of Personality and Social Psychology, 49, 95-1 12.

Reynolds, A. (1992). Comparing measures of parental involvement and their
effects on academic achievement. Early Childhood Research Quarterly, 7, 441-462.

Sarason, R. B., Sarason, I. G., & Pierce, G. R. (1990). Traditional views of social
support and their impact on assessment. In R. B. Sarason, I. G. Sarason, and G. R. Pierce
(Eds), Social support: an Interactional view (pp. 319-366). Oxford, England: Wiley and
Sons.

Scott-Jones, D. (1995). Parent-child interactions and school achievement. In B.
A. Ryan, G. R. Adams, T. P. Gullotta, R. P. Weissberg, & R. L. Hampton (Eds), The
Family-school connection (pp. 75-107). Thousand Oaks, CA: Sage.

Shumow, L. & Miller, J. D. (2001). Parents’ at-home and at-school academic
involvement with young adolescents. Journal of Early Adolescence, 21, 68-91.

138

Slaughter, D. T. & Epps, E. G. (1987). The Home environment and academic
achievement of Black American children and youth: An Overview. Journal of Negro
Education, 86, 3-20.

Smith, E. P., Atkins, J. & Connell, C. M. (2003). Family, school, and community
factors and relationships to racial-ethnic attitudes and academic achievement. American
Journal of Community Psychology, 32, 159-173.

Soodak, L. C. & Erwin, E. J. (2000). Valued member or tolerated participant:
Parents’ experience in inclusive early childhood settings. Journal for the Assocaition for
Persons with Severe Handicaps, 25, 29-41.

Swap, S. M. (1990). Comparing three philosophies of home-school collaboration.
Equity and Choice, 6, 9-19.

Tardy, C. (1985). Social support measurement. American Journal of Community
Psychology, 13, 187-202.

Thoits, P. A. (1982a). Conceptual, methodological and theoretical problems in
studying social support as a buffer against life stress. Journal of Health and Social
Behavior, 23, 145-159.

Thoits, P. A. (1982b). Life stress, social support, and psychological vulnerability:
Epidemiological considerations. Journal of Community Psychology, 10, 341-362.

Thompson, F. T. (2002). Student achievement, selected environmental
characteristics, and neighborhood types. The Urban Review, 34, 277-292.

US. Department of Education, National Center for Education Statistics, National

Household Education Survey. (1999). Parent and Family Involvement in Education:
1996 (NCES 1999-001). Washington, DC: US. Government Printing Office.

US. Department of Education, National Center for Education Statistics. (2002).
The Condition of Education 2002, NCES 2002-025, Washington, DC: US. Government
Printing Office.

US. Department of Education, National Center for Education Statistics. (2003).
The Condition of Education 2003, NCES 2003-067, Washington, DC: US. Government
Printing Office.

Vaden-Kieman, N. & McManus, J. (2005). Parent and Family Involvement in
Education: 2002-03 (NCES 2005-043). US. Department of Education, National Center
for Education Statistics. Washington, DC: US. Government Printing Office.

Vaux, A. (1988). Social support: theory, research and interventions. New York:
Praeger.

139

Wenz-Gross, M., Siperstein, G. N., Untch, A. S. & Widaman, K. F. (1997).
Stress, social support, and adjustment of adolescents in middle school. Journal of Early
Adolescence, 17, 129-151.

Whitcher, S. J. & Fisher, J. D. (1979). Multidimensional reaction to therapeutic
touch in a hospital setting. Journal of Personality and Social Psychology, 3 7, 87-96.

Wirt, J ., Choy, S., Rooney, S., Sen, A., & Tobin, R. (2004). The condition of
education 2004, NCES 2004-077.

Ysseldyke, J ., Dawson, P., Lehr, C., Reschly, D., Reynolds, M., & Telzrow, C.
(1997). School Psychology: A Blueprint for Training and Practice II. US. Maryland:
National Association of School Psychologists.

Zigler, E. F. & Gilman, E. (1991). Beyond Academic Instruction: the Twenty-

first-century school model for preschoolers. New Directions for Child Development, 53,
75-82.

140

  
 

          

 
  

     
  
 

  

\llllfllllllllllllllljl

02 56

TTTTTTT

\llllllllllljlllllllll

319