139
071

 

_THS

 

awe ‘ UBRARY
Michigan State
University

 

 

 

This is to certify that the
dissertation entitled

THE OPTIMAL MATCHING
OF PERCEIVED TEACHER SOCIAL SUPPORT
TO STUDENTS’ DAILY SCHOOL STRESSORS

presented by

Laura Jane Dilly

has been accepted towards fulfillment
of the requirements for the

PhD. degree in School Psychology

 

 

3 Major Professor’s Signature
{/(p / o 5

Date

MSU is an Affinnative Action/Equal Opportunity Institution

 

 

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

 

DATE DUE

DATE DUE

DATE DUE

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

2/05 W IR Judd-p.15

7*

THE OPTIMAL MATCHING OF PERCEIVED TEACHER SOCIAL SUPPORT TO
STUDENTS’ DAILY SCHOOL STRESSORS

By

Laura Jane Dilly

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

2005

ABSTRACT

THE OPTIMAL MATCHING OF PERCEIVED TEACHER SOCIAL SUPPORT TO
STUDENTS’ DAILY SCHOOL STRESSORS

By

Laura Jane Dilly

Students’ subjective experiences of daily stressors in school can negatively impact their
development and wellness. However, the social support that teachers provide for
students can be an important factor in promoting children’s wellness and buffering the
effects of stress. This study examined the optimal matching theory, which states that
specific types of social support are differentially effective at buffering the effects of
specific stressors, within a sample of upper elementary students. Students completed
self—report measures of daily school stressors, perceived teacher social support, academic
self-concept, and classroom climate. It was predicted that certain types of support
(Emotional and Instrumental) would buffer the effects of academic stressors and social
stressors on academic self-competence and peer self-esteem. Results did not support the
optimal matching theory. Teacher provided social support buffered the effects of daily
stressors on peer self-esteem. In addition, emotional support was a significant predictor
of academic self-concept, as were emotional support and instrumental support for peer

self-esteem.

ACKNOWLEDGEMENTS

I gratefully acknowledge my committee members Dr. Jean Baker, Dr. John
Carlson, Dr. Tom Luster, and Dr. Darrell Meece for sharing their expertise, thoughtful
critiques, and collaboration. In particular, I would like to thank Dr. Jean Baker, my
advisor and chairperson, for her continual support, insightful suggestions, and mentorship
through the years of graduate school and dissertation process.

I extend thanks to the Spencer Foundation and the committee for the Michigan
State University Continuing Education Fellowship which provided funding for the
preliminary study for my dissertation.

In addition, I would like to thank the Haslett Public School District’s
administration, teachers, and parents who collaborated with me and allowed me to collect
data in their schools. I greatly appreciate the efforts of the 3”, 4m, and 5th grade students
who completed the survey, without whom this dissertation would not be possible.

Finally, I would like to sincerely thank my friends and family. I appreciate your
patience through the many ups and downs involved in writing this dissertation and

faithful encouragement over the past five years.

iii

TABLE OF CONTENTS

LIST OF TABLES ............................................................................... v
LIST OF FIGURES .............................................................................. vi
INTRODUCTION ............................................................................... 1
LITERATURE REVIEW ....................................................................... 4
METHODS ....................................................................................... 24
RESULTS ......................................................................................... 29
DISCUSSSION ................................................................................... 34
BIBLIOGRAPHY ................................................................................ 44
APPENDICES .................................................................................... 52

iv

LIST OF TABLES

. Race and Sex Demographics of Participants ............................................. 50

. Scale Reliabilities ............................................................................. 51

. Descriptive Statistics: Means and Standard Deviations by Sex ........................ 53
. Descriptive Statistics: Means and Standard Deviations by Grade ..................... 55
. Inter-correlations between Variables ....................................................... 57
. Summary of Regression Analyses Predicting Academic Self-Competence ........... 58

. Summary of Regression Analyses Predicting Peer Self-Esteem ........................ 59

LIST OF FIGURES

1. Graph of Interaction of Stress and Social Support on Peer Self-Esteem ............... 60

vi

Introduction

There is an increasing interest in positive psychology, or the psychology of
wellness (Cowen, 1994). The field of positive psychology emphasizes factors that
enhance development and positive adjustment in contrast to deficit orientations that
emphasize the study of maladjustment (Albee, 2000; Cowen, 2000). This focus on
adjustment offers a more hopeful perspective on children’s development. As the
processes involved in supporting children’s development are better understood, children’s
environments can be tailored to support development and prevent future problems.

Within positive psychology, a number of conceptual models explaining
developmental phenomena have arisen, namely resilience, competencies, and stress and
coping models (Compas, Connor-Smith, Altzman, Thomsen, & Wadsworth, 2001;
Masten & Coatsworth, 1998). All of the models are similar, in that they concern
children’s adjustment. However, there are some distinctions between them. The term
resilience refers to the developmental process of positive life adjustment despite stressors
that place a child at risk (Doll & Lyon, 1998; Luthar, Ciccheti, & Becker, 2000). The
literature concerning the stress and coping model is situated within the developmental
psychopathology literature and emphasizes the process of adaptation. In contrast to the
resilience and stress and coping models, the competencies based model focuses solely on
the characteristics of an individual or environment that encourage positive adaptation.
Compas and colleagues (2001) summarizes the relationships between these models
stating, “Coping can be viewed as efforts to enact or mobilize competence or personal
resources, and resilience can be viewed as the successful outcome of these actions” (p.

89).

In order to encourage children’s positive development, the processes involved in
supporting their development must be further examined. According to an ecological
systems perspective, children develop within multiple, interacting contexts
(Bronfenbrenner, 1992). As schools provide a significant context in which children
develop, examining and understanding the proCesses that support children’s development
within schools is important. One of the most frequently cited protective factors, or
processes supporting development, for children is a relationship with a caring adult and
the social support provided within these relationships (Doll & Lyon, 1998; Masten &
Coatsworth, 1998). Tietjen (1994) emphasizes that social support exists within
relationships in various contexts. In addition, the importance of sources of social support
may differ developmentally, with children relying on adults and adolescents relying
increasingly more on peers (Gore & Aseltine, 1995). Therefore, elementary children
likely rely heavily on adult-provided social support. Within schools, the social support
that students receive within the teacher-student relationship may be of particular
importance to students’ positive development.

Rutter (1987) points out the importance of viewing social support not as a
variable but as a process. Therefore, it is important to further understand how this process
unfolds. Currently, within the resilience literature there is a call for greater attention to
the psychological processes and mechanisms responsible for the positive adjustment of
children and adolescents (Luthar et al., 2000). There have been a number of criticisms of
the lack of explanations of how or why a particular process is effective in facilitating the
adjustment of children within current literature on resilience (Pianta & Walsh, 1998;

Rutter, 1987). At the same time, within the stress and coping literature, a discussion

about the importance of increasing the specificity of models is taking place (Cicchetti &
Cohen, 1995; McMahon, Grant, Compas, Thurrn, & By, 2003). Grant and his colleagues
(2003) propose that “there is specificity in relations among particular stressors,
moderators, mediators, and psychological outcomes” (p. 453). Therefore, understanding
the types of teacher provided social support that are helpful within different situations can
begin to answer some of these questions.

The purpose of this study was to add specificity to the theoretical model of stress
and coping within the context of the daily stressors children experience in school.
Specifically, this study tested the optimal matching theory that states that for specific
stressors, specific types of support are most helpful. Teacher provided social support was
considered as an important mechanism involved in children’s experiences of coping with

daily school stressors.

Review of Literature

Major Theories of Social Sumrt

There are several major theoretical perspectives of social support, namely the
attachment model, the main effect model, and the stress buffering model. The first
theoretical perspective of social support focuses on the personality-based aspects of
social support. An individual’s attachment style is one factor that is proposed to affect
social support. According to perspectives of social support that draw on attachment
theory, children’s early relationships with their parents form enduring relational schemas
that contribute to their ability to elicit and perceive support from others (Pierce, Baldwin,
& Lydon, 1997). Therefore, social support is seen as an enduring personality
characteristic. Children develop support schemas in the context of relationships, with
their first relationship with their parents being particularly important. Support schemas
influence an individual’s expectations about the social environment’s willingness and
ability to meet future support needs (Pierce, Sarason, Sarason, Joseph, & Henderson,
1996). The support schemas that children develop in the context of their relationship with
their parents may be carried over into their relationship with a teacher (Pianta, 1994).
Pianta, Nimetz, and Bennett (1997) found that preschoolers’ teacher-student relationship
was related to their mother-child relationship, hypothesizing that this was due to
relational schema rather than intelligence or temperament. Students may perceive the
support that teachers provide differently based on their attachment styles.

There are two other major models of social support, the main-effect model and the
stress-buffering model. In contrast to the personality view of social support, these

models emphasize social support as an environmental provision. Therefore, these models

provide more direct implications for the provision of services to support children’s
development. Because the current study considers the importance of various types of
teacher provided social support, an environmental provision, it is more closely aligned
with this view of social support. Both the main-effect model and the stress-buffering
model have been empirically supported (Cohen & Wills, 1985). The main-effect model of
social support emphasizes the role of social support in promoting adjustment independent
of the existence of stress. Therefore, the presence of social support is assumed to be
important whether or not the individual is experiencing stress. For example, parental
social support may positively affect school attendance for all children, independent of
any stressors the child may be experiencing. The buffering model delineates a specific
relationship between stress and social support in the adjustment process. Social support
reduces the negative effect of the stressor on the adjustment process. For example,
parental social support may buffer the effects of poverty on school attendance.
Mechanisms involved in social support. Two mechanisms have been proposed to
explain the positive effects of social support. The first explanation focuses on
individuals’ cognitive appraisals and the second explanation is based on self-
determination theory. The focus on cognitive appraisal process explanation is grounded
in the stress and coping literature. The variations in people’s ability to cope with stress
are explained by the cognitive processes that occur between a person’s encounter with
stressors and their reaction (Lazarus & Folkman, 1984). In particular, an individual’s

cognitive appraisal, or evaluation of a situation, influences their reaction and ability to

cope.

Social support may influence individuals’ primary or secondary cognitive
appraisals (Cohen, Gottlieb, & Underwood, 2000; Sandler, Miller, Short, & Wolchik,
1989). Primary appraisal refers to the cognitive processes involved in evaluating the
significance of the situation for one’s well-being. The perception that others will provide
support may redefine the situation and reduce the perception of threat that the situation
presents. For example, a student may perceive that their teacher will help them complete
a math worksheet that is difficult. Therefore, the student may perceive less stress related
to the task. Secondary appraisal refers to the cognitive processes involved in evaluating
what can be done to mange threatening or challenging situations. Perceived or received
social support may reduce the negative cognitive, emotional, physiological, and/or
behavioral responses to a perceived stress. The social support “may alleviate these
negative reactions by providing a solution to a problem, reducing the importance of the
problem, or providing a distraction from the problem” (Cohen, Gottlieb, & Underwood,
2000, p. 14). For example, a student may be called a name by another student. A teacher
may emotionally support the student by expressing that they care about the child,
reducing the student’s negative affective reaction to the stressful event.

A second mechanism for understanding social support derives from theories
within human motivation. Self-detennination theory assumes there are three essential
psychological needs: relatedness, autonomy, and competence (Ryan, 1995). Sandler and
his colleagues (1989) highlight three intervening processes through which stress and
support may affect children’s adjustment, namely sense of security of social
relationships, perceptions of control, and self-esteem. Similarly, Ryan and Solky (1996)

propose that social support can meet individuals’ psychological needs, specifically the

needs for relatedness and autonomy. First, social support provided by teachers may
allow students to feel more connected and cared for, fulfilling the need for relatedness.
Osterman (2000) discusses the important role that teachers’ support and relationships
with students can play in fostering a sense of relatedness. Second, social support
provided by teachers encourages students’ autonomy, self-esteem, perceived competence,
and internal motivation (Deci, Schwartz, Sheinman, & Ryan, 1981; Ryan & Grolnick,
1986)
Stressors in School for Children

There are a number of definitions of stress and stressors in relation to children.
The most commonly cited definition of stress is the Lazarus and F olkrnan (1 984) one
stating, “Psychological stress involves a particular relationship between the person and
the environment that is appraised by the person as taxing or exceeding his or her
resources and endangering his or her well being” (p. 19). Grant and his colleagues
(2003) suggest that because individual’s cognitive appraisals of the stressfulness of
situations are often not studied within stress research, stress could be defined as
“environmental events or chronic conditions that objectively threaten the physical and/or
psychological health or well-being of individuals of a particular age in a particular
society” (p. 450). However, the current study is particularly interested in stress as
defined by the first definition, specifically, children’s subjective appraisals of
environmental situations as stressful.

While the bulk of literature examining childhood stressors has focused on chronic
or major stressors that children encounter such as poverty, chronic illness, parental

divorce, and abuse, children also experience more minor life stressors (Grant et al.,

2003). Minor life stressors, or daily hassles, can include normative experiences of
development such as being picked last for a team or having difficulty with a homework
assignment (D’Aurora & F imian, 1988; Elias, Ubriaco, Reese, Gara, Rothbaum, &
Haviland, 1992; Phillips, 1978). Within school, these daily hassles can exist within
several domains including academic, personal, and social situations (Burnett &
Fanshawe, 1997; Grannis, 1992; Phillips, 1978; Wenz-Gross, Siperstein, Untch, &
Widaman, 1997).

Several studies have examined the frequency of school stressors and students’
cognitive appraisals of the events. Phillips (1978) measured the stressful events in
elementary students’ experience of school. Overall, achievement stressors were greater
than social stressors for these students. Sources of achievement stress included the
teacher speaking too quickly, meeting teacher expectations, making a mistake while in
front of the class, and taking tests. Sources of social stress included being teased by peers
and being unpopular with peers. Grannis (1992) found that middle school students
reported three domains of stressors: academic troubles, physical and personal assaults,
and general school disrupters. General school disrupters were the most frequent
stressors, followed by academic troubles, and physical and personal assaults. However,
students reported that they were most upset by physical and personal stressors, followed
by academic and general school disrupters. Within a group of middle school students,
Elias and his colleagues (1992) found that daily stresses clustered around five areas:
adaptation difficulty, substance abuse, peer relationships, conflict with authority, and
academic pressures. Stressors related to adaptation, authorities, and peers were appraised

as causing more distress than stressors related to academic pressures and substance abuse.

This is similar to the Grannis (1992) study in that social stressors were particularly
distressing for middle school students.

There are relatively few studies that consider the perceptions of stressors between
students of different genders. Few gender differences have been found in perceived
number of stressors encountered in school (Elias et al., 1992; Phillips, 1978). However,
Phillips (1978) found a difference in magnitude, with girls appraising stressors as being
more stressful. Therefore while boys and girls may perceive the same number of
stressors, girls may appraise these same stressors as more stressful.

Children in middle childhood experience a variety of situations which they
appraise as exceeding their resources. These daily stressors exist within academic, social,
and personal domains. Knowledge of children’s subjective experience of these stressors
provides an important step in understanding children’s coping; however, the processes
involved in children’s coping must also be explored.

Multidimensional Aspects of Socigl Support

Social support is often viewed as a broad set of processes that can be broken
down into functional types of social support. Each of these types of social support is
hypothesized to serve a different function. There have been numerous attempts to
categorize the different types of social support. Cobb (1976) conceived social support as
information provided by others which fall into three classes. First, the information may
provide emotional support, making an individual feel that they are cared for and loved.
Second, the information may provide esteem support, encouraging an individual to feel
that they are of value and worth. Third, the information may provide an individual with

the sense that they belong to a social network that shares mutual obligation.

House (1981) surveyed the types of social support discussed within the literature
and proposed four major types of supportive behavior or acts, namely emotional support,
instrumental support, informational support, and appraisal support. Emotional support
refers to the provision of empathy, care, and love. Instrumental support refers to
behaviors that directly help an individual such as physical and monetary aid.
Informational support refers to providing an individual with information to aid them in
coping with personal and environmental difficulties. Finally, appraisal support refers to
the provision of evaluative information related to an individual’s worth.

Cutrona and Russell (1990) reviewed five multidimensional models of social
support and found empirical support for five major functional dimensions of social
support: emotional support, esteem support, tangible aid, information support, network
support, and opportunity to provide nurturance to others. In this model, esteem support
and tangible aid correspond to the House (1981) definitions of appraisal support and
instrumental support, respectively. Network support refers to a person’s sense that they
part of a group of people that share common interests. The opportunity to provide
nurturance to others was added as a sixth functional aspect of social support; however,
the authors noted that the mechanisms involved might significantly differ from other
types of social support.

Cutrona and Suhr (1994) divided the five major dimensions of social support into
two broad categories: action-facilitating support and nurturant support. Action-
facilitating support aims to help a stressed individual solve or eliminate the problem
causing distress. It can include both informational and tangible support. In contrast,

nurturant support does not aim to solve the problem but instead focuses on comforting the

10

individual. Nurturant support includes emotional support and network support. Esteem
support can take the form of either action-facilitating or nurturant support. If the esteem
support focuses on assuring the individual that they have the abilities and skills to solve
the problem, this may support action. However, if the esteem support focuses on
assuring the individual that they have value and worth in order to lessen the intensity of a
distressed individual’s negative emotions, this is more consistent with nurturant support.

Within the literature regarding children, Malecki, Demaray, and Elliott (2000)
have adopted a four factor model of social support following the House (1981) social
support typologies. The model features informational, emotional, instrumental, and
appraisal support. In addition, Richman and colleagues (1998) proposed an 8 factor
model of support children encounter including. They eight types of support were
technical challenge, emotional challenge, listening, reality confirmation, personal
assistance, confirmation, and technical appreciation.
Perceived and received social support

The social support literature distinguishes between available support and enacted
social support (Tardy, 1985). Perceived support is the support that an individual
subjectively feels is available or has been provided in the past for them. Received
support examines the objective enactment of support. Dunkel-Schetter and Bennett
(1990) reviewed empirical studies that included measures of both available and received
support and found evidence for a conceptual distinction between the two. In the review
of eight studies, correlations between measures of available and received support ranged

from no relationship to a moderate relationship.

11

Overall, perceived support has more consistently been shown to be related to
lower levels of psychological symptomatology than received support (Wills & Shinar,
2000). Perceived social support may be particularly important because the positive effects
of social support are only possible when an individual appraises the support as available
(House, 1981). Cohen and Wills (1985) add that it may be the perceived adequacy rather
than the perceived availability of social support that is particularly important. Support
perceived as available may not be judged as adequate and therefore not positively affect
an individual. Wills and Shinar (2000) suggest that since perceived support measures
have more extensive empirical support, the use of perceived support measures in studies
is a solid, conservative choice. However, they encourage the simultaneous use of both
perceived and received support to gain a better understanding of how the two constructs
relate.

Some gender differences have been reported in perceived social support. Females
tend to report significantly higher levels of perceived total support as well as emotional
and informational support (Demaray & Malecki, 2002a; Dubow and Ullman, 1989;
Malecki & Elliot, 1999). Jackson and Warren (2000) add that females are more likely to
seek social support than males after exposure to stress. Therefore, females may perceive
higher levels of social support because they seek out higher levels of social support.
Sources of Supmrt

Children can receive social support from a number of different sources including
parents, peers, close friends, and teachers. A number of studies have shown that children
are able to distinguish among sources of support (Dubow & Ullman, 1989; Malecki &

Demaray; 2002). In addition, children who tend to perceive higher levels of social

12

support fiom one source, tend to perceive higher levels of social support from other
sources (Dubow & Ullman, 1989). Demaray and Elliot (2001) found that students’
perceptions of support and their parents’ and teachers’ reported provision of support were
moderately related. The importance of the various sources of support may differ also
across a child’s development. For example, the support from peers may become
increasingly important as children move from middle childhood to adolescence (Malecki
& Demaray, 2002; Malecki & Elliot, 1999).
Efficacy of Social Supmrt

The efficacy of social support in buffering the effects of stress has been measured
in a number of different ways. Several studies have measured the effects of social
support on behavioral and social outcomes, such as behavioral problems, substance
abuse, clinical adjustment, interpersonal adjustment, grades, and attendance (Demaray &
Malecki, 2002b; Jackson & Warren, 2000; Lifrak, McKay, Rostain, Alterman, &
O’Brien, 1997; Rosenfeld, Richman, & Bowen, 2000). However, outcome measures of
subjective well-being may be of particular importance since the experience of stress is a
subjective experience. In addition, measures of subjective well-being are also consistent
with positive psychology in that they measure health rather than negative outcomes.
Various studies have found that social support buffers the effects of stress on measures of
subjective well-being, such as school satisfaction, self-concept, and self-efficacy
(Rosenfeld, Richman, & Bowen, 2000; Wenz-Gross et al., 1997).
Empirical studieLof social support

Several studies have empirically considered the relationship between perceived

social support and outcomes for children and adolescents. Students who report receiving

13

lower levels of social support from their parents, friends, and teachers have poorer school
outcomes than students who report higher levels of social support (Rosenfeld et al.,
2000). While students who perceive support from teachers, parents, and peers show
higher positive student outcomes including higher grades, school satisfaction,
engagement, and self-efficacy (Rosenfeld et al., 2000).

Different sources of social support may be more important for different outcomes.
Lifrak et al. (1997) found that higher levels of perceived teacher support, and to a lesser
degree parent support, were significantly associated with lower levels of cigarette,
alcohol, and marijuana use for middle school boys. For girls, higher levels of perceived
peer support were significantly associated with lower levels of cigarette and marijuana
use. Garnefski and Diekstra (1996) found that for high school students, negative
perceptions of school support related to increased behavioral problems, negative
perceptions of peer support was related to increased emotional problems, and negative
perceptions of family support were associated with emotional and behavioral
dysfunction. Parent, teacher, and school support has also been shown to be negatively
correlated to school outcomes while parent and classmate support was related to clinical
and interpersonal indicators (Demaray & Malecki, 2002b). Wenz—Gross et a1. (1997)
found that family support was more related to academic self-concept, whereas peer
support was more related to social self-concept. Richman, Rosenfeld, and Bowen (1998)
found teacher support was significantly related to two middle school outcomes, school
satisfaction and school self-efficacy, as well as two high school outcomes, sense of
coherence and time spent studying. Attendance, avoidance of problem behavior, grades,

prosocial behavior, school satisfaction, school self-efficacy, sense of coherence, and time

14

spent studying were also related to adult caretaker and peer support. Therefore, teacher
provided social support may play a particularly important role in school related
outcomes.

The types of support that students perceive from their parents, teachers, peers, and
friends differ. In a group of middle school students, informational support was most
highly reported from teachers (Malecki & Demaray, 2003). In addition, emotional and
information support were the most highly reported types of support from parents and
emotional and instrumental support scores were the most highly reported from peers and
friends.

Various types of social support have also been shown to be related to different
outcomes. Richman and his colleagues (1998) found that different types of social support
were related to different types of positive outcomes for at-risk students. Attendance was
related to technical challenge and emotional challenge support; avoidance of problem
behavior was related to technical challenge support; grades were related to listening
support and reality confirmation support; prosocial behavior was related to personal
assistance support; school satisfaction was related to emotional support, emotional
challenge support, and reality confirmation support; school self-efficacy was related to
confirmation support, listening support, and emotional support; sense of coherence was
related to technical appreciation support and emotional support; and time spent studying
was related to technical appreciation support, emotional support, emotional challenge
support, and personal assistance support. Wenz-Gross and colleagues (1997) found that
emotional support from family members moderated the relationship between peer stress

and feelings of depression for adolescent students. In addition, problem-solving support

15

from adults other than parents (potential school staff) moderated the effect of
teacher/rules stress on liking school.

The sufficiency and necessity of teacher social support for positive student
outcomes is currently unclear. Rosenfeld and colleagues (2000) found that support from
teachers was a necessary condition for many school outcomes but not a sufficient
condition. While teacher support alone was not sufficient for positive school outcomes in
this study, teacher support in combination with support from parents or peers was
necessary. However, Malecki and Demaray (2003) report that emotional support
perceived from teachers was the sole significant predictor of students’ social skills and
academic competence. In this sample, teacher support also predicted school
maladjustment and parent support predicted personal adj ustrnent.

Matching Theog

There are multiple models of stress and coping; however, this study is focused on
the optimal matching theory. The optimal matching theory adds specificity to the stress
and coping model, matching support to specific stressors. Jackson (1992) outlines two
major versions of the matching hypothesis. The first suggests that the source of social
support should be matched to the stressor and assumes that the source of support should
match the source of the stressor. For example, work stressors would best be buffered by
sources of support at work (House, 1981). Therefore, stressors at school could be
potentially best buffered by sources of support at school, such as support provided by a
teacher.

Cutrona and Russell (1990) propose a second version of the matching theory

stating that “different kinds of social support are most useful in the context of different

16

kinds of stressful life events” (Cutrona & Suhr, 1992, p.168). Therefore, specific types of
social support may differentially lead to more positive outcomes for individuals,
depending upon the stressor. In addition, Cutrona and colleagues argued that the most
beneficial type of support is related to the controllability of the stressor (Cutrona, 1990;
Cutrona & Russell, 1990). When stressors are uncontrollable, there is nothing that can be
done to lessen the consequences of the stressor, so the most important task for the
individual is to recover from the negative emotions related to the stressor. However,
when a stressor is controllable, action can be taken to decrease the occurrence or
consequences of the stressor. Therefore, it was proposed that stressors that are
uncontrollable require emotional support whereas stressors that are controllable require
problem-focused support or instrumental support (Cutrona, 1990; Cutrona & Russell,
1990)

Studies of optimal matching within adult populations. Several studies have
examined the optimal matching hypothesis with adult and college age pOpulations.
Cutrona and Suhr (1994) examined the controllability hypothesis with thirty married
couples. One member of the couple discussed a stressful situation with their spouse. The
interaction was coded for the frequency of support behaviors, type of support behaviors,
and controllability of the stressful event. The spouses disclosing the stressful event also
rated their satisfaction with the interaction. A significant positive correlation was found
between the coded controllability of the stressful event and the number of information
support behaviors offered by the spouse. In addition, when the stressful event was judged
to be controllable by the support giver, information support positively predicted the

support receiver’s satisfaction. When the stressful event was judged to be controllable by

17

the support receiver, information support negatively predicted the support receiver’s
satisfaction. In a second study, each member of the partnership was asked to rate his or
her ability to control the event. In this study, a significant positive correlation was found
between the support receiver’s perceived controllability and the number of esteem
support behaviors offered by the spouse. Therefore, when the support receiver perceives
that they can decrease the occurrence or consequences of the stressor they are
experiencing, their partner may respond with encouragement of capabilities and skills. In
both studies, regardless of the controllability of the stressful event by either partner,
emotional support was significantly related to higher satisfaction with the interaction.

Horowitz and colleagues (2001) further defined the optimal matching theory of
social support to consider the type of problem experienced by the speaker, the speaker’s
goals, and the listener’s provision of different types of support. Each of these factors
within the interpersonal relationship can be considered with regards to the dimensions of
communion and agency. The content of problems is defined as either agentic or
communal. Agentic problems focused on action and implied a specific goal. Communal
problems focused on the internal distress of an individual and imply a need to feel more
connected and loved. For example, performance failures are agentic problems whereas
romantic losses are communal problems. The speaker’s goal is the type of support the
speaker is attempting to elicit, either agentic or communal support. The listener can
provide either agentic, problem-solving focused, or communal, empathetic focused,
responses.

Horowitz and his colleagues (2001) conducted a series of studies to test aspects of

the optimal-matching hypothesis. In the first study, college undergraduates were asked to

18

read a problem scenario such as a romantic loss or a performance failure. Then they were
asked to imagine the scenario had just happened to them and they were going to talk to a
close friend about the problem. They were then given a list of 19 possible responses that
their fliend could provide and rate the extent to which each response would make them
feel worse or better using a Likert-type scale. The 19 responses loaded onto two factors
corresponding to communal reactions and agentic reactions.

Horowitz and colleagues (2001) found that agentic problems evoked agentic
support, whereas communal problems evoked communal support. In this study,
undergraduates read stories focused on an academic performance failure, an agentic
problem, and a breakup in a romantic relationship, a communal problem. Each
participant was then instructed to write how they would provide support to their fliend
with each problem. The responses were coded for the type of response given. Overall,
the undergraduates produced more agentic reactions to all stories. Results showed
significant main effect for the type of reaction and a significant interaction effect.
Performance failure scenarios generated relatively greater numbers of agentic reactions,
whereas romantic breakup scenarios generated relatively greater numbers of communal
reactions.

The way that a support seeker flames a request influences the type of support
provided (Horowitz et al., 2001). Agentic flames containing statements such as “I don’t
know what to do” were found to elicit significantly more agentic support whereas
communal flames contained statements such as “I feel terrible” were found to elicit
significantly more communal support. However, the magnitude of the flame effect was

greater for communal problems. In addition, the support receiver and support provider

19

were both more satisfied with an interaction when the type of support elicited matched
the type of support provided. For example, both the support receiver and the support
provider were more satisfied if the receiver elicited communal support and the provider
gave communal support.

Chen and Tang (1997) examined the optimal matching hypothesis within the
context of the stresses and social support experienced by Chinese mothers of adult
children with mental retardation. Overall, the mothers rated tangible support as more
effective than emotional and informational support. In addition, for controllable
stressors, tangible support was rated as more effective than informational support.

Finally, Tetzloff and Barrera (1987) tested the matching hypothesis within a
group of divorcing mothers. The moderating effect of various types of social support on
the relationship between various stressors and negative psychological outcomes was
examined. Stressors included tangible stress, parenting stress, and social stress while the
types of social support included parenting support, emotional support, and tangible
support. Limited support was found for the matching hypothesis; however, the authors
mention a number of methodological limitations that made detection of significant
moderators difficult. Included were issues related to a lack of subjects for sufficient
power and the use of support and stress typologies that had limited empirical support.

Therefore, several factors may influence the optimal match between stressors and
support. First, an individual’s appraisal of the controllability may influence the optimal
type of support. When an individual judges the event as controllable, esteem support
may be optimal. Whereas for events judged as uncontrollable, informational support may

be optimal. Second, the type of stressor may influence the optimal match. For stressors

20

that involve a specific goal, problem-solving or informational support may be optimal.
However, for stressors that are focused on an individual’s internal distress, communal or
emotional support may be optimal. Third, groups of individuals, such as mothers of
children with disabilities, may experience unique stressors for which specific types of
social support are most helpful. Finally, within studies of the optimal matching theory,
carefiil attention must be paid to statistical issues related to power.

Studies of optimal matching hypothesis with child populations. There is currently a
dearth of empirical evidence considering the optimal matching hypothesis within child
populations. However, several studies have begun to test the matching hypothesis. Gore
and Aseltine (1995) considered the match between various stressors and differing sources
of social support. Family support was found to buffer the effects of family stressors on
adolescent mood. Similarly, peer support was found to buffer the effects of peer stressors
on mood. Therefore, the authors concluded that family and peer domains are fairly
distinct during adolescence, causing stressors within each domain to best be buffered by
supports within the same domain. For example, family stressors are best buffered by
support within the family. Wenz-Gross and colleagues (1997) tested several models to
consider whether differing types of stress and differing sources and types of social
support were related to adjustment outcomes for middle school students. For academic
self-concept, academics and peer relationships were sources of stress while family
emotional support was an important source of support. For social acceptance, self-
concept within the peer domain and teachers/rules were sources of stress while peer

companionship support was an important source of support. For liking of school,

21

teachers/rules was a stressor and family emotional support was an important source of

support.

Pre-dissertation Study

A pre-dissertation, exploratory study was conducted to explore 3”, 4th, and 5th
grade students’ experiences of daily school stressors and teacher provided social support.
UCHRIS approval was obtained prior to the start of the study. Parental consent and child
assent was obtained. A total of 5 students currently residing at the Veterans of Foreign
Wars National Home for Children were interviewed. Semi-structured interviews (Rubin
& Rubin, 1995) focused on: a) the daily hassles that elementary students experience in
school and b) the social support that teachers provide in these situations. A copy of the
interview protocol is contained in Appendix A. Interviews were transcribed and
analyzed. A cross case analysis was completed and as themes appeared, excerpts were
labeled and organized into files, creating thematic connections (Merriam, 1998). This
study helped to confirm the school stressors and support constructs to be utilized within
the current study. Children identified academic and social stressors. In addition, they
described examples of emotional, instrumental, informational, and appraisal support.
Summa_ry

Children experience a number of daily stressors in schools. These stressors can
involve both academic and social demands within the school environment that
overburden children’s resources. However, the social support that children receive flom
adults, including teachers, can buffer the potentially negative effects of the daily school
stressors that students encounter and encourage positive development. Teachers can

provide a variety of types of social support for students including instructional,

22

emotional, instrumental, and appraisal support. Of particular importance are children’s
subjective experiences of stress and social support, as these are more tied to outcomes
than objectively rated stressors or received social support. Currently, our understanding
of the buffering effects of social support lacks specificity. However, the optimal
matching theory of social support adds specificity to the stress-buffering hypotheses,
stating that for specific types of stressors, specific types of social support are more
effective in buffering stress. Empirical studies of the optimal matching theory have
shown supporting evidence for the model within adult populations and initial support
within child populations.
Pmse of the Study

The positive psychology literature has consistently emphasized the protective
effects of children’s relationships with caring adults. Therefore, understanding the role
that teachers play in providing social support for children, will further our understanding
of this protective process. Specifically, the focus on this study was to test the optimal
matching hypothesis by examining the relationship between various daily stressors and
the type of teacher provided support of elementary students. This study addresses the
question: Are different types of teacher provided social support differentially helpful in
buffering the effects of different school stressors on positive school outcomes?

The hypotheses are based on self-determination theory. It is hypothesized that

for academic stressors that may threaten the students’ sense of competence, instrumental
support will better buffer the effects of academic stressors on academic self-concept.

Secondly, it is hypothesized that for social stressors that may threaten the students’ sense

23

of relatedness, emotional support will better buffer the effects of social stressors on peer

self-esteem.

24

Methods
Participants

The sample of third through fifth graders was drawn flom two elementary schools
in a Midwestern school district. A total of 134 students were recruited for participation,
68 flom school one and 66 flom school two. The school sent home permission forms to
eligible students. Within this sample there were 71 boys and 63 girls. The participants
were in grades three (11: 42), four (n=36), and five (n=56) and ranged in age flom 8 to 12
years old. The sample included 42 3rd grade students, 36 4th grade students, and 56 5th
grade students.

A total of 115 were White, 4 Asian, 2 Hispanic/Latino/Latina, 1 African
American, 11 multi-racial, and 1 student omitted racial background. Further
demographic information is presented in Table 1. During the 2002-2003 school year, the
racial makeup of the district was 86% White, 5% Asian, 4% Hispanic/Latino/Latina, and
5% Aflican American (National Center for Education Statistics, 2005). Information on
students with multi-racial backgrounds was not available. White students composed 85%
of study sample and composed 86% of the district population. Overall, 14% of the
students in the district’s population were flom racial minority backgrounds and 13% of
the students in the study sample were flom racial minorities when students flom multi-
racial backgrounds are included in this percentage.

Procedure
Active informed consent was required for participation in the study. Permission

forms were distributed to eligible students by the school and were returned to the school

25

by the students. The overall participation rate was 22%, with 23% at school 1 and 22% at
school 2.

Data were collected during the spring quarter. Students completed a battery of
self-report measures designed to assess perceived schools stressors, social support,
academic self-concept, and peer self-esteem. The questionnaires were administered in
small groups, ranging in size flom 4-12 by trained university researchers. Protocols were
read aloud to students to control for reading ability and administered in a counter-
balanced order between groups to control for order effects.

Measures

School stressors. Daily school stressors were measured using two subscales of
the Things That Happen in School (THIS; Grannis, 1992). The THIS consists of 15
items. Two aspects of daily school stresses were measured using child ratings:
Academic Troubles, which measures stressors related to academic aspects of school, and
Physical and Personal Assaults, which measures stressors related to interpersonal aspects
of school. The Physical and Personal Assaults subscale was modified to include an
additional 5 items taken flom the Orpinas (1995) Victimization scale (Appendix C) in
order to increase the validity of the construct. Children were asked to rate both the
flequency of the items’ occurrence and the stress appraisal. Frequency ratings consisted
of a 3-point Likert-type scale, with scores ranging flom I (hardly ever) to 3 (often).
Stress appraisal ratings consist of a 5-point Likert-type scale, with scores ranging flom 1
(very good) to 5 (very upset). In previous studies, internal consistency was reported to be
.79 for Stressor Frequency and .87 for Stress Appraisal (Grannis, 1992). Within this

sample, the internal consistency was .85 and .88 for the total Stress Frequency and Stress

26

Appraisal scales, respectively. The reliabilities for the Academic Stressors Frequency
and Stress Appraisal Scales were .63 and .68, respectively. The reliabilities for the
modified Physical and Personal Assaults Frequency and Stress Appraisal scales were .86
and .88, respectively.

Teacher social supmrt. Perceived teacher social support was measured using the
Teacher subscale of the Child and Adolescent Social Support Scalp (Appendix D;
Malecki, Demaray, & Elliot, 2000). This Teacher subscale measures 4 types of social
support: emotional, informational, appraisal, and instrumental. The scale consists of 12
items, 3 items for each type of support. Students rated both the flequency and
importance of each teacher supportive behavior. Frequency ratings consists of a 6-point
Likert-type scale, with scores ranging flom 1 (never) to 6 (always). Importance ratings
consisted of a 3-point Likert-type scale, with scores ranging flom 1 (not important) to 3
(very important). The factor structure and reliability of this scale are well established; in
previous studies the internal consistency for the type scores range flom .81 to .82
(Malecki & Demaray, 2003). For this sample, the reliability of the entire Frequency scale
was .86 and the Importance scale was .80. Within this study, the four social support
subscales were collapsed to form two social support scales due to low initial reliabilities
of the four individual subscales. Emotional and appraisal support were collapsed to form
a new emotional support variable. Informational and instrumental support were collapsed
to form a new instrumental support variable. The reliabilities of subscales are reported in
Table 2.

Academic self commtence. Self-concept was assessed using a modified version

of the Scholastic Competence subscale of the Self-Perception Profile for Children

27

(Appendix E; Harter, 1985). The subscale used a 4-point Likert-type scale, with scores
ranging flom 1 (never) to 4 (almost always). The Scholastic Competence consists of 6
items that measure children’s perceptions of their ability within the realm of scholastic
performance. Internal consistency reliabilities for the original scale range flom .80 to .85
(Harter, 1985). Within this study, the internal consistency for the modified scale was .75.

Peer self-esteem. Children’s self-esteem in relation to peers was measured with
the Peer Subscale of the Self-Esteem Questionnaire (Appendix F; DuBois, Felner, Brand,
Phillips, & Lease, 1996). The 8 item subscale used a 4-point Likert-type scale, with
scores ranging flom 1 (never) to 4 (almost always). The internal consistency reliabilities
for the scale range flom .85 to .86 (DuBois et al., 1996). Within this study, the internal
reliability was .80.

W

Descriptive statistics were used to describe the sample and measures. The
internal consistency of each measure was calculated. In addition, Pearson product-
moment correlations between variables were calculated.

Analyses examined potential differences by grade and gender on total stressors,
total social support, scholastic competence, and social acceptance. AN OVAs were
utilized for analyses involving potential grade differences and t-tests for gender
differences.

Analyses to test the optimal matching theory followed a model described by
Tetzloff and Barrera (1987). Regression analyses consisted of predictors entered in the
following order: stress appraisal scale, social support importance scale, and the product

of the stress and support scale. A significant interaction term indicates a moderating

28

effect of social support (Holmbeck, 1997). A moderator is “a qualitative (e.g., sex, race,
class) or quantitative (e.g., level of reward) variable that affects the direction and/or
strength of the relation between an independent or predictor variable and a dependent or
criterion variable” (Barron & Kenny, 1986, p. 1174). Therefore, social support may
change the relationship between daily school stressors and children’s self-concept if it
acts as a moderator. Support for the optimal matching hypothesis would be indicated if
one interaction term (e.g., academic stress x emotional support) was significantly greater
than other potential stress x type of social support interactions (e.g., academic stress x
instrumental support).

Two analyses were run, one for each of the dependent variables (academic self-
competence and peer self-esteem). For scholastic competence, academic stressors,
emotional support, instrumental support, and interaction terms (academic stressors x
emotional support and academic stressors x instrumental support) were entered. For peer
self-esteem, social stressors, emotional support, instrumental support, and interaction
terms (social stressors x emotional support and social stressors x instrumental support)
were entered. In order to eliminate problems of multicollinearity between the two main
effect terms and the interaction term, the independent variables, scholastic competence
and peer social acceptance, and the moderators, types of social support, were centered.
Sufficient statistical power to detect medium effect sizes was expected with the proposed

number of subjects and the proposed analyses (Green, 1991).

29

Results

Means and standard deviations for each of the variables are presented in Table 3
and Table 4. Because the scores reflecting total stress appraisal, academic stress
appraisal, social stress appraisal, total support flequency, were based on the combination
of individual scores, scaling procedures were employed to assess the reliability or internal
consistency of these summed scores. The reliability coefficients are reported in Table 2.
Due to the low reliabilities for the individual teacher social support scales, emotional and
appraisal support were combined to form a new emotional support scale and
informational and instrumental support were combined to form a new instrumental
support scale. The formation of these two scales is consistent with a number of studies
that have failed to identify several types of social support, but instead find a more clear
distinction between emotional support and instrumental support (Cutrona & Suhr, 1994;
Rook & Underwood, 2000).

Intercorrelations are presented in Table 5. Correlations between total support and
total stressors were significant (r = .291, p<.01). Correlations between total support,
emotional support, and instrumental support were significant (r = .479, p<.01; r =.444,
p<.01; r = .439, p.01). However, correlations between measures of stress and the .
outcome measures, academic self-competence and peer self-esteem, were not significant.
The two outcome measures, academic self-competence and peer self-esteem, were also
significantly correlated (r = .352, p<.01). The magnitudes of the correlations were not

indicative of a multicollinearity problem.

30

Independent t-test analyses were calculated to determine if there were group
differences across gender on the total stress appraisal, academic stress appraisal, social
stress appraisal, total social support flequency, emotional support flequency, instrumental
support flequency, scholastic competence, and peer self-esteem. Cases with missing data
were excluded analysis by analysis. Results revealed that there were no significant
differences for gender on social stress appraisal, total social support flequency, emotional
support flequency, instrumental support flequency, scholastic competence, and peer self-
esteem. Results revealed significant differences for gender on total stress appraisal (t 132
= -2.09, p = .04, d = .36) and academic stress appraisal (1132 = -2.35, p = .02, d = .46),
with females reporting higher levels of stress.

AN OVA analyses were conducted to determine if there were group differences
across grade on the total stress appraisal, academic stress appraisal, social stress
appraisal, total social support frequency, emotional support flequency, instrumental
support flequency, scholastic competence, and peer self-esteem. Cases with missing data
were excluded analysis by analysis. Overall, the amount of missing data was negligible.
Results showed no significant differences for grade on academic stress appraisal,
instrumental support flequency, scholastic competence, and peer self-esteem. Results
revealed significant differences for grade on social stress appraisal (F 2,131= 3.845, p =
.024, Eta-squared = .02), total stress appraisal (F 2,131: 3.097, p = .049, Eta squared =
.05), and total support flequency (F 2,131= 3.172, p = .045, Eta squared = .05). Tukey’s
post-hoe tests were used to further examine differences by grade. On social stress
appraisal, 4th graders reported significantly higher levels than 5’h graders (p = .019, d =

.60). On total stress appraisal, 4th graders reported significantly higher levels than 5’h

31

graders (p = .03 7, d = .56). On total support flequency, 4th graders reported significantly
higher levels than 5th graders (p = .036, d = .55). Since moderate effect sizes were
detected for gender and grade, gender and grade were treated as potential covariates
within regression equations and entered as the first two steps of the regression equations.
Dummy codes were created for gender and grade. In order to code for grade, two
separate dummy variables were created.

Data was visually examined and diagnostic tests were conducted to assess for
violations of regression assumptions. Regression analyses were then completed to assess
interaction effects. Missing data were eliminated pairwise.

In order to test the first hypothesis, which stated that for academic stressors
instrumental support will better buffer the effects of academic stressors on academic self-
concept, a multiple regression was used. The regression analysis assessed the interaction
between academic stress and emotional support as well as the interaction between
academic stress and instrumental support with academic self-competence as the
dependent variable. Gender was entered in the first step, grade was entered in the second
step, and academic stressors, emotional support, instrumental support, academic stressors
x emotional support, and academic stressors x instrumental support were entered in the
third step. The first two steps which included gender (F(l,132) = .03, p=.854) and grade
F(3,130) = .46, p=.714) were not significant. Therefore, gender and grade were removed
and the regression was rerun with the remaining variables. The F for the overall
regression model was significant F(5,128) = 8.13, p<.001, and 32 = .241, adjusted 132 =

.211. Emotional support ([3 = .511, p<.001) was a significant predictor of academic self-

32

competence. Academic stressors, instrumental support, and the interaction terms were
not significant. Table 6 summarizes the results, including standardized [3 values.

In order to test the second hypothesis, which stated that for social stressors
emotional support will better buffer the effects of social stressors on peer self-esteem, a
multiple regression analysis was used. The regression analysis assessed the interaction
between social stress and emotional support as well as the interaction between social
stress and instrumental support with peer self-esteem as the dependent variable. Gender
was entered in the first step, grade was entered in the second step, and social stressors,
emotional support, instrumental support, social stress x emotional support, and social
stressors x instrumental support were entered in the third step. The first two steps which
included gender (F(1,132) = 1.07, p=.303) and grade F(3,130) = 1.34, p=.263) were not
significant. Therefore, gender and grade were removed and the regression was rerun with
the remaining variables. The F for the regression model was significant F(5 ,128) = 9.48,
p<.001, and 132 = .270, adjusted 32 = .241. Emotional support (13 = .347, p = .002) and
instrumental support ([3 = .220, p = .041) were significant predictors of peer self-esteem.
Social stressors and the interaction terms were not significant. Table 7 summarizes the
results.

In order to assess the general stress-buffering hypothesis, two regressions were
then conducted to assess the interaction of total stressors and total support on both
dependent variables, academic self-competence and peer self-esteem. Tables 6 and 7
summarize these results. Gender was entered in the first step, grade was entered in the
second step, and total stressors, total support, and total stressors x total support were

entered as the independent variables and academic self-esteem were entered in the third

33

step. The first two steps for gender (F( 1,132) = .03, p=.854) and grade F (3,130) = .46,
p=.714) were not significant. Therefore, gender and grade were removed and the
regression was rerun with the remaining variables. The F for the regression was
significant F(3,130) = 8.18, p<.001, and 1:2 = .159, adjusted 32 = .139. Total support ([3 =
.389, p <.001) was a significant predictor. Total stressors and the interaction term were
not significant.

For the second analyses, gender was entered in the first step, grade was entered in
the second step, and total stressors, total support, and total stressors x total support were
entered in the third step. Peer self-esteem was entered as the dependent variable. The
first two steps for gender (F (1,132) = .1.07, p=.303) and grade F(3,130) = .1.34, p=.263)
were not significant. Therefore, gender and grade were removed and the regression was
rerun with the remaining variables. The F for the regression was significant F(3,130)
=15.46, p<.001, and 32 = .263, adjusted 32 = .246. Total Support ([3 = .523, p <.001) and
Total Stressors x Total Support ([3 = .154, p = .046) were significant predictors.

In order to examine the interaction effect, the interaction was plotted following
procedures suggested by Aiken and West (1991). The regression equation was recast as
the regression of the criterion, peer self-esteem, on one predictor, total support. This
regression equation was Y = (.23 + .01(Stress))Support + (-.05(Stress) + (-.17)). Low,
medium, and high values of total stress were calculated. These values corresponded to
one standard deviation below the mean of stress, the mean of stress, and one standard
deviation above the mean of stress. These values of stress were then substituted into the
equation above, creating a series of three simple regression equations. Plots of these

regression lines revealed a cross-over interaction effect. The plot is shown in Figure 1.

34

On inspection of the interaction graph, it appears that when stress is moderate, social
support functions differently. Under conditions of medium stress, social support is less

predictive of peer self-esteem than when stress is low or high.

35

Discussion

Teacher provided social support is an important factor leading to children’s
positive adjustment. The overall amount of support provided by children’s teachers
predicts both their academic self-competence and peer self-esteem. This is consistent
with other studies that have found that teacher provided social support is a significant
predictor of academic and social competencies (Malecki & Demaray, 2003; Rosenfeld et
al., 2000). In particular, the emotional support provided by teachers predicts children’s
academic self-competence and peer self-esteem. Teacher’s instrumental support is also
predictive of students’ peer self-esteem.

The study failed to find support for the optimal matching hypothesis. There was
no evidence that specific forms of social support were more effective in buffering the
effects of academic and social stressors on academic self-competence and peer self-
esteem. The interactions between specific stressors and particular forms of social support
were not significant. Several other studies examining the optimal matching hypothesis
have also failed to support the hypothesis. At this time, there is no conclusive support for
the optimal matching theory in the child or adult literature. In a review of empirical
research, Cutrona and Russell (1990) found two-thirds of studies added support to the
optimal matching hypothesis, while one-third did not.

An individual’s perceived controllability of a stressor may dictate the type of
social support that is most effective in buffering the effects of the stressor (Cutrona,
1990; Cutrona & Russell, 1990). For stressors perceived as uncontrollable, emotional
support is most beneficial whereas for stressors perceived as controllable, instrumental

support is most beneficial. Within this study, while emotional support was not shown to

36

buffer the effects of stressors, emotional support was a significant predictor of both
academic self-competence and peer self-esteem. Therefore, emotional support may be
particularly important for children’s development. According to the optimal matching
theory, this may be because children perceive that they have less control within their
environment and therefore benefit flom emotional support more than instrumental
support. The provision of emotional support may further encourage their sense of self-
determination, including their sense that that they are competent and able to make
changes within their environment.

There are several plausible reasons for the lack of support of the matching
hypothesis. First, students may vary too much individually for there to be a match
between a specific stressor and a specific type of social support within a population.
While an individual student may consistently appraise specific stressors as more stressful
and find specific types of teacher support most helpfirl, there may not be patterns of
matching within an entire population of students. However, there may be specificity of
stressors and support for specific groups of children. For example, children with
difficulties learning versus those with behavioral difficulties may require different teacher
provided support. Baker, Davis, Dilly, Aupperlee, & Patil (2002) found that students at
low, moderate, and high levels of risk differed in the teacher provided social support that
predicted school satisfaction. In addition, specificity of stressors and support by gender
may exist. The current study found higher levels of both perceived stress and perceived
support for girls. This is consistent with previous research (Demaray & Malecki, 2002a;
Dubow and Ullman, 1989; Malecki & Elliot, 1999; Phillips, 1978). These differences

may be attributable to differences in girls’ perceptions of stress and support or they may

37

be attributable to objective differences in the flequency of stressors and support girls
receive. This study did not support differing specificity patterns for boys and girls;
however, larger samples may find that specificity patterns may vary by gender.

A second explanation for the failure to find support for the optimal matching
theory may be that daily stressors rarely affect a child in only one way. Rook and
Underwood (2000) note that social interactions often have multiple, simultaneous goals.
Therefore, teachers may be responding to stressors by providing multiple forms of social
support to facilitate the child’s coping with a single stressor that affects the child in
multiple ways. For example, a student who does not understand an assignment may
benefit flom informational support, (i.e. illustrating how to complete the assignment) as
well as emotional support (i.e. encouraging them to continue). The preliminary study for
the current study found that 3-5 grade students spontaneously recalled multiple types of
support teachers provided for a single academic or social stressor. For example, one
student described that when she did not understand schoolwork, an academic stressor, she
thought the teacher’s emotional support (i.e. stating good job) and instrumental support
(i.e. explaining the instructions better) would be helpful.

A third potential reason for the failure to support the matching hypothesis may be
that the ftmction of teacher provided social support may not be best explained by a stress
and coping model. The stress and coping model assumes that social support buffers the
effects of existing stressors. However, social support may serve a preventative role,
decreasing the likelihood of stressful events, and/or a counteractive role, directly
increasing the variables leading to adjustment but being unrelated to stress (Sandler, et

al., 1989). Teacher provided social support may be best explained by a main-effects

38

model which is consistent with a counteractive role. The main-effects model proposes
that the presence of social support promotes the positive adjustment independent of the
existence of stress. Social support acts to increase positive adjustment at all levels of
stress. In this study, social support acted similarly for students reporting low and high
levels of stress, predicting higher levels of academic self-competence and peer self-
esteem. Luthar, Cicchetti, and Becker (2000) suggest referring to attributes with direct
arneliorative effects at high and low levels risk conditions as “protective” factors.
Viewing social support flom this perspective, as a protective factor, is consistent with
positive psychology and resilience literature because it emphasizes building student
wellness. Further, self-determination theory, emphasizing the need for belonging,
autonomy, and competence, may explain the mechanism involved in relationship between
teacher social support and increased positive outcomes. Perceived social support
increases students’ feelings of belonging, autonomy, and competence (Deci etal., 1981;
Ryan & Grolnick, 1986), which may reduce students’ perceptions of future events as
stressors.

Surprisingly, the correlations were not significant between measures of stress and
the two outcomes, academic self-competence and peer self-esteem. While the main
effects model could explain this finding, the lack of correlation may also be due to some
measurement problems. On the measure of stress, the inclusion of some items that may
not be true daily stressors such as “I was warned that I might not pass a subject or pass to
the next grade” may have affected this correlation. However while correlations were not
significant between measures of stress and the two outcomes; correlations were

significant between measures of support and the two outcomes. Attachment theory offers

39

one possible reason for the moderate correlation between measures of perceived teacher
social support and peer self-esteem. According to attachment theory, students form
enduring relational schemas that influence their expectations within all of their
relationships. Therefore, students who perceive a strong, supportive relationship between
themselves and their teachers may be more likely to also perceive a positive relationship
between themselves and their peers.

The significant interaction of stress and social support on peer self-esteem
signifies that for various levels of stress, social support functions differently on peer self-
esteem. Inspection of the interaction graph indicates that when stress is moderate, social
support functions differently. Students perceiving moderate levels of stress and low
teacher support, had the highest peer self-esteem. There are several possible
explanations. First, students with moderate levels of stress may have an optimal
facilitative level of stress. Anxiety performance research has shown that individuals
perform best when there is a moderate level of stress (Kirkland, 1971; Yerkes & Dodson,
1908). At low levels of stress, students may not be aroused enough to perform at higher
levels; however, at high levels of stress students have difficulty functioning. Therefore
moderate levels of daily stressors may encourage students’ development, including their
sense of peer self-esteem. Students perceiving moderate levels of stress and low teacher
social support may rely more heavily on peer support, increasing their interactions with
peers and possibly providing opportunities to increase their peer self-esteem.

Limitations

Several methodological limitations may have contributed to finding non-

significant moderating effects. In general, moderating effects are extremely difficult to

40

detect in field based research due to several factors (McClelland & Judd, 1993). First,
while the power was estimated to be at moderate, additional subjects within the study
could have increased the statistical power, increasing the likely of detecting a moderating
effect. In addition, McClelland and Judd (1993) also suggest over sampling extreme
cases increase the likelihood of detecting a moderating effect. Second, the measure of
academic stressors had an acceptable but low reliability ((1 = .63). Therefore, this
increased the general error within the regression equations and made detecting a true
moderating effect more difficult. Third, the moderately high correlation between the
emotional support and instrumental support may have masked some effects. Fourth,
combining emotional, instrumental, informational, and appraisal support into two broad
areas of support may have obscured matching effects.

Several theoretical limitations may influence the study. In addition to the
measure of academic daily stressors having a low reliability, this measure also may not
have fully captured the construct of academic daily stress. For example, it included fairly
inflequent daily stressors such as failing a grade while neglecting to include stressors
such as not understanding class work. Second, in the preliminary study, several students
mentioned environmental changes, including changes to the task or physical
environment, that teachers make within the classroom to support them. For example,
students stated that teachers can support them by shortening a spelling test or moving
their desk. While the measure of teacher provided social support used in this study did
measure instrumental support which is similar to environmental support, there were not
any specific items that measured environmental support flom teachers such as the

students described in the preliminary study.

41

Finally, the relatively low participation rate, 22%, may have affected the
representativeness of the study sample. Research within schools that utilizes active
parental consent typically reach participation rates of 30-60% (Esbensen, Miller, Taylor,
He, & F reng, 1999; Tigges, 2003). The rates within this study were slightly lower than
these estimates. Children whose parents provide active consent tend to be white,
younger, females, exhibit more prosocial behavior, and be flom intact families (Esbensen
et al., 1999; Tigges, 2003). Therefore, this study may have underrepresented students
experiencing greater daily stressors.

Implications for Practice

This study holds several implications for practice. First, this study found teacher
social support was a better predictor of students’ positive outcomes than the daily
stressors that they encounter. Therefore, it may be more important to pay attention to
providing social support within classrooms than exclusively focusing on reducing risks
for students. This emphasizes a more proactive approach, focused on increasing student
wellness. This proactive focus on the social-emotional wellness of students is consistent
with positive psychology. There are several universal prevention programs that
emphasize the development of strong teacher-student relationships and supportive
classroom environments. For example, the Responsive Classroom developed by the
Northwest Foundation for Children is an approach to creating a positive classroom
environment for academic and social-emotional learning (Northeast Foundation for
Children, 2005).

In addition, support can be provided for teachers to enable them to increase the

social support that they provide for students. Cutrona and Cole (2000) suggest a number

42

of techniques for increasing support within a natural support network. These suggestions
can be adapted to specifically address the support teachers provide for students.
Therefore, the support teachers provide for students could potentially be increased by 1)
increasing teachers’ understanding of the specific stressors students encounter, 2)
modeling, rehearsing, and providing feedback for teachers regarding the behaviors
involved in providing support for students, 3) changing teachers’ attitudes that interfere
with the provision of support, such as feeling that providing social support is not central
to their role as teacher, 4) coordinating the staff within the school to provide specific
support responsibilities and avoid overwhelming the classroom teacher, and 5) offering
support to teachers to help them sustain their efforts.
Future research

This study holds several implications for future research related to teacher
provided social support and its impact on positive student adj ustrnent. While support for
the optimal matching theory was not found in this study, optimal matching theories
should not be abandoned. Specificity could provide important information in meeting the
needs of students and maximizing positive adjustment. Future students may benefit flom
considering how social support acts for various student groups such as students with
learning difficulties and behavioral problems. Considering entire student populations
may mask specificity that exists within smaller subgroups of students.

Alternatively, future research may focus on considering teacher provided social
support flom a main effects model. Teacher provided social support may function
independently of stress and may benefit children at any level of stress. Further,

differentiating how social support functions to promote positive adjustment of children

43

would be beneficial in developing specificity to social support models and in developing
wellness in children.

In addition, while this study focused on students’ subjective experiences of stress
and support, future studies may examine received support through the use of
observational data or provided support as measured by teacher report. Comparisons of
perceived and received social support could be made. These comparisons may illuminate
differences in how students or perhaps different groups of students perceive teacher
provided social support.

Summm

This study tested the optimal matching theory by examining the relationship
between various daily stressors and the type of teacher provided support of elementary
students. No support was found for the optimal matching theory. Teacher provided
social support was a significant predictor of both academic self-competence and peer
self-esteem, emphasizing the importance that teachers play in supporting students’
positive adjustment. In addition, some evidence was found for the buffering effects of
teacher provided social support on the relationship between daily school stressors and
peer self-esteem. As schools serve as an influential environment for children’s
development and teachers as important adults within children’s lives, continued efforts to
understand the functioning of social support is essential to further develop children’s

positive adjustment within schools.

44

APPENDICES

45

Appendix A: Social Support Interview Protocol

Intro: I am a student at Michigan State University and I want to find out some more
about what it is like to be a [3”, 4’”, or 5“] grader. To find out more, I am interviewing
some experts like you. So you ’11 need to put on your expert cap and think real hard to
help me out. 0k? If there are any questions that I ask that don ’t make sense, just tell
me?

Demographic information
1) Age?

2) Grade Level?

3) Sex?

Student Perceptions of Supmrt flom Teachers
1) What types of things make a teacher helpfirl?

Student Perceptions of Stressors in School (Refer to expert status again.)
1) What kinds of things stress out kids in school?

Alternative: What kinds of problems do kids your age have in school?
What kinds of things bother kids in school?
What kind of hassles do kids have in school?
Annoy? Bother?

Probes:
1. Having trouble with fliends? (i.e. getting in a fight on the
playground)
11. Having trouble with other people in the class?
111. Having trouble with school work? (ie. not having enough time?, when the

teacher is telling you what to do?, forgetting homework, during reading?,
during math?, not having materials?)

IV. Projects?

V. Having trouble with teachers? (teacher getting mad, getting in trouble)

2) How much does this bother kids?

Etcher Social Support

1) What things do teachers do to help kids with this problem? (Pose in terms of
situations generated by student. Complete questions 1-6 for entire situation and then
move to next generated stressful situation.)

Probes:

Do teachers ever talk to kids to help them? What do they say?

What would you see teachers doing if they were helping someone?

Do teachers ever give kids things to help them? What do they give them?

What do teachers do if it isn’t that big of a problem?

46

2) What could teachers do that would be more helpful (more supportive)?
Alternative: Are there other things that teachers could do to help kids?
Probe: Talking to kids (emotional support)?
Helping kids solve the problem (instructional support)?
Giving kids more materials or more time (tangible support)?
Helping kids realize that it is not that big of a problem (appraisal support)?

3) How helpful are the things teachers do?
Probe:
- What is the most helpful thing a teacher could do?

4) If the teacher did that, how would kids feel?

Conclusion of Interview
1) Is there anything else you think I should know?

 

47

Appendix B

Things that Happpn at School (Grannis, 1992)

Academic Troubles Subscale

1. I couldn’t get an answer to a question I asked.

2. I was warned that I might not pass a subject or pass to the next grade.
3. I or one of my friends was moved to a different class.

4. Someone tried to pressure me to skip class.

5. I could not finish my work in class.

6. I didn’t have a pen or paper or my book to do my work in class.
7. I did worse on a test than I should have.

8. There was a change of teachers in one or more of my classes.

Physical and Personal Assaults Subscale

1. Someone tried to copy my work when I didn’t want them to.
2. Someone hit me, or kicked or pushed me.

3. Someone made fun of something I said in class.

4. Someone shouted or screamed at me.

5. Somebody treated me like I always do wrong.

6. Someone was picking on me in the hall.

7. Someone took something flom me.

Appendix C

Items flom Victimization Scale (Orpinas, 1995)

l. A student teased me to make me angry.

2. A student said things about me to make other students laugh.
3. A student called me bad names.

4. A student tried to hurt my feelings.

5. Other students did not want to spend recess or lunch with me.

Appendix D

Child and Adolescent Social Support Sgafi (Malecki, Demaray, & Elliot, 2000)
Teacher Subscale

Emotional Support

1. My teacher cares about me.

2. My teacher treats me fairly.

3. My teacher makes it okay to ask questions.

Informational Support

1. My teacher explains things that I don’t understand.

2. My teacher shows me how to do things.

3. My teacher helps me solve problems by giving me information.

Appraisal Support

1. My teacher tells me I did a good job when I’ve done something well.
2. My teacher nicely tells me when I make mistakes.

48

3. My teacher tells me how well I do on tasks.

Instrumental Support

1. My teacher makes sure I have what I need for school.

2. My teacher takes time to help me learn to do something well.
3. My teacher spends time with me when I need help.

Appendix E

Self-Perception Profile for Children (Harter, 1985)
Scholastic Competence

I am very good at my school work.

I am smart enough to do my school work.

I am slow in finishing my school work.*

I often forget what I learn.*

I do very well at my school work.

I can figure out the answers to school work.

999959!”

* Reverse keyed.

Appendix F

Self-Esteem Oucstionnpir_e (DuBois, F elner, Brand, Phillips, & Lease, 1996)
Peers Subscale

1. I am as popular with kids my own age as I want to be.

2. I am as good as I want to be at making new fliends.

3. I have a many close friends as I would like to have.

4. I am as well liked by other kids as I want to be.

5. I feel good about how well I get along with other kids.

6. I wish my fliends liked me more than they do.

7. I feel good about how much my friends like my ideas.

8. I feel OK about how much other kids like doing things with me.

49

Table 1

Race and Sex Demographics of Participants

 

 

 

Race
Caucasian Aflican Hispanic/ Asian/ Multi- Missing Total
American Latino/ Pacific Racial
Sex Latina Islander
Girls 52 1 1 2 6 1 63
Boys 63 0 1 2 5 O 71
Total 115 l 2 4 11 1 134

 

50

Table 2

 

 

Scale Reliabilities.
Variable Alpha Coefficient
Total Stressors- Appraisal .89
Academic Stressors — Appraisal .68
Social Stressors — Appraisal .88
Total Stressors- Frequency .85
Academic Stressors — Frequency .63
Social Stressors — Frequency .86

Total Teacher Social Support — Frequency .86

Emotional - Frequency .74
Informational — Frequency .55
Appraisal — Frequency .77
Instrumental — Frequency .72
Emotional” — Frequency .83
Instrumental" — Frequency .78

Total Teacher Social Support — Importance .80

Emotional — Importance .50
Informational — Importance .32
Appraisal — Importance .60
Instrumental — Importance .58
Emotional“ — Importance .69

51

Instrumental“ * — Importance .78
Scholastic Competence .80
Peer Self-Esteem .75
* Combined scale — Emotional and Appraisal Support

** Combined scale - Informational and Instrumental Support

52

Table 3

Descriptive Statistics: Means and Standard Deviation by Sex

 

 

 

Full Sample Male Female
N = 134 n = 71 n = 63

Variables M SD M SD M SD
Total Stressors- Appraisal 80.84 8.95 79.34 9.12 82.54 8.51

Academic Stressors — Appraisal 30.84 3.66 30.15 3.89 31.62 2.23

Social Stressors — Appraisal 50.00 6.14 49.18 6.18 50.92 6.01
Total Stressors- Frequency 37.76 8.42 37.94 8.48 37.57 8.44

Academic Stressors - Frequency 13.63 2.98 13.63 3.01 13.62 2.96

Social Stressors — Frequency 24.14 6.70 24.42 6.55 23.84 6.90
Total Teacher Social Support — 55.34 10.66 54.86 9.93 55.87 11.49
Frequency

Emotional — Frequency 15.12 2.78 14.96 2.78 15.30 2.80

Informational - Frequency 13.77 2.90 13.76 2.82 13.78 3.02

Appraisal - Frequency 13.79 3.46 13.80 3.21 13.78 3.74

Instrumental — Frequency 12.66 3.75 12.34 3 .44 13.02 4.07

Emotional“ — Frequency 28.91 5.63 28.76 5.37 29.08 5.94

Instrumental" - Frequency 26.43 5.94 26.10 5.66 26.79 6.27
Total Teacher Social Support - 28.01 4.21 27.29 4.13 28.88 4.18
Importance

Emotional — Importance 7.20 1.26 7.06 1.22 7.37 1.30

53

Informational — Importance
Appraisal — Importance
Instrumental - Importance
Emotional“ - Importance
Instrumental" - Importance
Scholastic Competence

Peer Self-Esteem

* Combined scale — Emotional and Appraisal Support

7.13

6.69

6.93

13.90

14.06

20.24

25.34

1.17

1.57

1.45

2.44

2.24

2.86

4.66

7.04

6.45

6.79

13.51

13.83

20.28

24.94

** Combined scale — Informational and Instrumental Support

54

1.20

1 .47

1.52

2.24

2.32

2.81

4.61

7.24

6.97

7.08

14.33

14.32

20.19

25.78

1.13

1.65

1.36

2.59

2.12

2.92

4.71

Table 4

Descriptive Statistics: Means and Standard Deviation by Grade

 

 

 

3rd Grade 4m Grade 5’’1 Grade
n = 42 n = 56 n = 36

Variables M SD M SD M SD
Total Stressors- Appraisal 80.95 9.75 82.63 8.01 77.94 8.85

Academic Stressors — Appraisal 30.67 3.82 31.38 3.80 30.22 3.18

Social Stressors — Appraisal 50.29 6.53 51.25 5.394 47.72 6.28
Total Stressors— Frequency 39.13 8.10 36.74 8.89 37.79 8.05

Academic Stressors — Frequency 13.98 2.75 13.30 3.12 13.72 3.04

Social Stressors — Frequency 25.10 6.73 23.49 6.86 24.03 6.50
Total Teacher Social Support — 54.93 11.26 57.71 10.05 52.11 10.23
Frequency

Emotional — Frequency 15.45 2.60 15.27 2.66 14.50 3.13

Informational — Frequency 13.24 3.09 14.23 2.89 13.67 2.64

Appraisal - Frequency 13.67 3.69 14.68 3.13 12.56 3.34

Instrumental — Frequency 12.57 3.91 13.54 3.77 11.39 3.21

Emotional“ — Frequency 29.12 5.89 29.95 4.91 27.06 6.05

Instrumental" — Frequency 25.81 6.25 27.77 6.07 25.06 5.01
Total Teacher Social Support — 28.00 4.22 28.75 4.22 26.85 4.05
Importance

Emotional -— Importance 7.12 1.27 7.43 1.14 6.94 1.39

55

Informational — Importance
Appraisal — Importance
Instrumental — Importance
Emotional“ - Importance
Instrumental" — Importance
Scholastic Competence

Peer Self-Esteem

* Combined scale — Emotional and Appraisal Support

7.31

6.55

6.93

13.67

14.24

20.36

24.29

1.16

1.63

1.30

2.46

2.14

2.98

4.57

7.18

7.02

7.05

14.45

14.23

20.45

25.80

** Combined scale — Informational and Instrumental Support

56

1.25

1.51

1.54

2.32

2.42

2.50

4.52

6.86

6.36

6.72

13.31

13.58

19.78

25.83

1.02

1.55

1.49

2.48

2.03

3.24

4.90

Table 5

Inter-correlations between Variables

 

 

 

Variables

Variables 1 2 3 4 5 6 7 8
1. Total Stress — -- .853** .950** .291** .313** .227** .140 .042
Appraisal
2. Academic Stress- ’ .648** .253** .265“ .202* .163 .044
Appraisal
3. Social Stress- .274** .299** 210* .107 .034
Appraisal
4. Total Support— .917** .926" .398” .479“
Frequency
5. Emotional Support .699M .463** .444“
— Frequency
6. Instrumental .275** .439“
Support — Frequency
7. Scholastic .352"

Competence
8. Peer Self-Esteem
*p<.05

**p<.01

57

Table 6

Summary of Regression Analyses Predicting Academic Self-competence

 

 

Variable Adjusted F(dt) p value [3 p value
R2

Model 1 .211 8.13m”) <.001

Academic Stressors -.002 .978

Emotional Support .511 <.001

Instrumental Support -.090 .425

Academic Stressors x -.070 .549

Emotional Support

Academic Stressors x 1.661 .099

Instrumental Support

Model 2 .139 8.1 80,130) <.001

Total Stress .025 .767
Total Support 4.608 <.001
Stress x Support -.011 .894

58

Table 7

Summary of Regression Analyses Predicting Peer Self-esteem

 

 

Variable Adjusted F(do p value [3 p value
R2

Model 1 .241 9.48m”) <.001

Social Stressors -.089 .284

Emotional Support .347 .002

Instrumental Support .220 .041

Social Stressors x .046 .743

Emotional Support

Social Stressors x .138 .295

Instrumental Support

Model 2 .246 15.46(3,130) <.001

Total Stress -.088 .271
Total Support .523 <.001
Stress x Support .154 .046

59

Figure 1. Graph of Interaction of Stress and Social Support on Peer Self-Esteem.

 

Stress
4.00-
A High Struts
0 Medium Stress
0 Low Stress
2.001

p
8
l

cted peer self-esteem

"' 4.004

 

4.00 -

 

 

 

-10'.68 obo 10.88
Teacher Social Support

BIBLIOGRAPHY

Aiken, L.S. & West, S.G. (1991). Multiple regression: Testing and interpreting
interactions. Thousand Oaks, CA: Sage Publications.

Albee, G.W. (2000). The Boulder Model’s fatal flaw. American Psychologist, 55,
247-248.

Baker, J .A., Davis, S., Dilly, L., Aupperlee, J. & Patil, S. (2002). The
Developmental Context of School Satisfaction: Schools as Psychologically Healthy
Environments. Paper presented at the annual meeting of the American Psychological
Association.

Baron, R.M. & Kenny, DA. (1986). The moderator-mediator variable distinction
in social psychological research: Conceptual, strategic, and statistical considerations.
Journal of Personality and Social Psychology, 51, 1 173-1 182.

Bronfenbrenner, U. (1992). Ecological systems theory. In R. Vasta (Ed.), Six
Theories of Child Development: Revised Formulations and Current Theories, (pp. 187-
249). Bristol, PA: Jessica Kingsley Publishers, Ltd.

Burnett, P.C. & Fanshawe, J.P. (1997). Measuring school-related stressors in
adolescents. Journal of Youth and Adolescence, 26, 415-428.

Chen, T.Y. & Tang, CS. (1997). Stress appraisal and social support of Chinese
mothers of adult children with mental retardation. American Journal of Mental
Retardation, 101 , 473-482.

Cicchetti, D. & Cohen, DJ. (1995). Perspectives on developmental
psychopathology. In D. Cicchetti & D.J. Cohen (Eds), Developmental psychopathology:
Theory and methods, Vol. I (pp.3-20). New York: John Wiley & Sons.

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

Cohen, S., Gottlieb, B.H., & Underwood, LG. (2000). Theoretical and historical
perspectives. In S. Cohen, L.G. Underwood, & B.H. Gottlieb (Eds), Social Support
Measurement and Intervention: A Guide for Health and Social Scientists (pp. 3-28).
New York: Oxford University Press.

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

Compas, B.E., Connor, Smith, J .K., Saltzrnan, H., Thomsen, Alexandra Harding,
and Wadsworth, ME. (2001). Coping with stress during childhood and adolescence:

61

Problems, progress, and potential in theory and research. Psychological Bulletin, 127,
87-127 .

Cowen, EL. (1994). The enhancement of psychological wellness: Challenges
and opportunities. American Journal of Community Psychology, 22, 149-179.

Cowen, EL. (2000). Now that we all know that primary prevention in mental health
is great, what is it? Journal of Community Psychology, 28, 5-16.

Cutrona, CE. (1990). Stress and social support- In search of optimal matching.
Journal of Social and Clinical Psychology, 9, 3-14.

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

Cutrona, C.E. & Suhr, J .A. (1994). Social support communication in the context
of marriage: An analysis of couples’ supportive interactions. In HR. Burleson & T.L.
Albrecht (Eds.), Communication of Social Support: Messages, Interactions,
Relationships, and Community (pp. 113-135). Thousand Oaks, Ca: Sage Publications.

D’Aurora, D.L. & Fimian, M.J. (1988). Dimensions of life and school stress
experienced by young people. Psychology in the Schools, 25, 44-53.

Deci, E.L., Schwartz, A.J., Sheinman, L., & Ryan, RM. (1981). An instrument
to assess adults’ orientations toward control versus autonomy with children: Reflections
on intrinsic motivation and perceived competence. Journal of Educational Psychology,
73, 642-650.

Demaray, M.K. & Elliot, SN. (2001). Perceived social support by children with
characteristics of attention-deficit/hyperactivity disorder. School Psychology Quarterly,
16, 68-90.

Demaray, M.K. & Malecki, C.K. (2002a). Critical levels of perceived social
support associated with student adjustment. School Psychology Quarterly, 17, 213-241.

Demaray, M.K. & Malecki, C.K. (2002b). The relationship between perceived
social support and maladjustment for students at risk. Psychology in the Schools, 39,
305-316.

Doll, B. & Lyon, MA. (1998). Risk and resilience: Implications for the delivery

of educational and mental health services in schools. School Psychology Review, 27,
348-363.

62

Dubow, E.F. & Ullman, DC. (1989). Assessing social support in elementary
school children: The survey of children’s social support. Journal of Clinical Child
Psychology, 18, 52-64.

Dunkel-Schetter, C. & Bennett, TL. (1990). In B.R. Sarason, I.G.R. Sarason, &
G. Pierce (Eds.), Social Support: An Interactional View (pp. 267-296). New York: John
Wiley & Sons.

Elias, M.J., Ubriaco, M., Reese, A.M., Gara, M.A., Rothbaum, P.A., Haviland, M.
(1992). A measures of adaptation to problematic academic and interpersonal tasks of
middle school. Journal of School Psychology, 30, 41-57.

Esbensen, P.A., Miller, M.H., Taylor, T.J., He, N., & F reng, A. (1999).
Differential attrition rates and active parental consent. Evaluation Review, 23, 316-335.

Garnefski, N. & Diekstra, R. (1996). Perceived social support flom family,
school, and peers: Relationship with emotional and behavioral problems among
adolescents. Journal of the American Academy of Child and Adolescent Psychiatry, 35,
1657-1 664.

Gore, S. & Aseltine, R.H. Jr. (1995). Protective processes in adolescence:
Matching stressors with social resources. American Journal of Community Psychology,
23, 301-327.

Grannis, J .C. (1992). Students’ stress, distress, and achievement in an urban
intermediate school. Journal of Early Adolescence, 12, 4-27.

Grant, K.E., Compas, B.E., Stuhlmacher, A.F., Thurm, A.E., McMahon, S.D., &
Halpert, J .A. Stressors and child and adolescent psychopathology: Moving flom
markers to mechanisms or risk. Psychological Bulletin, 129, 447-466.

Green, 8.3. (1991). How many subjects does it take to do a regression analysis?
Multivariate Behavioral Research, 26, 499-510.

Harter, S. (1985). Manual for the Self-Perception Profile for Children. Denver,
CO: University of Denver.

Holrnbeck, G.N. (1997). Toward terminological, conceptual, and statistical
clarity in the study of mediators and moderators: Examples flom the child-clinical and
pediatric psychology literature. Journal of Consulting and Clinical Psychology, 65, 599-

610.

Horowitz, L.M., Krasnoperova, E.N., Tatar, D.G., Hansen, M.B., Person, E.A.,
Galvin, K.L., & Nelson, KL. (2001). The way to console may depend on the goal:
Experimental studies of social support. Journal of Experimental Social Psychology, 37,
49-61.

63

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

Jackson, RB. (1992). Specifying the buffering hypothesis: Support, strain, and
depression. Social Psychology Quarterly, 55, 363-378.

Jackson, Y. & Warren, J .S. (2000). Appraisal, social support, and life events:
Predicting outcome behavior in school-age children. Child Development, 71 , 1441-1457.

Kirkland, MC. (1971). The effects of tests on students and schools. Review of
Educational Research, 41, 303-350.

Lazarus, R.S. & F olkman, S. (1984). Stress, appraisal, and coping. New York:
Springer Publishing Company.

Liflak, P.D., McKay, J.R., Rostain, A., Alterman, A.I., O’Brien, GP. (1997).
Relationship of perceived competencies, perceived social support, and gender to
substance use in young children. Journal of the American Academy of Child and
Adolescent Psychiatry, 36, 933-940.

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, MK. (2002). Measuring perceived social support:
Development of the Child and Adolescent Social Support Scale (CASSS). Psychology in
the Schools, 39, 1-18.

Malecki, C.K. & Demaray, MK. (2003). What type of support do they need?
Investigating student adjustment as related to emotional, informational, appraisal, and
instrumental support. School Psychology Quarterly, 18, 231-252.

Malecki, C.K., Demaray, M.K., & Elliott, SN. (2000). Child and Adolescent
Social Support Scale. DeKalb, IL: Northern Illinois University.

Malecki, C.K. & Elliot, SN. (1999). Adolescents’ ratings of perceived social
support and its importance: Validation of the student social support scale. Psychology in
the Schools, 36, 473-483.

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

McClelland, G.H. & Judd, CM. (1993). Statistical difficulties of detecting
interactions and moderator effects. Psychological Bulletin, 114, 376-390.

64

McMahon, S.D., Grant, K.E., Compas, B.E., Thurm, A.E., By, S. (2003). Stress
and psychopathology in children and adolescents: Is there evidence of specificity?
Journal of Child Psychology and Psychiatry, 44, 107-133.

Merriam, SB. (2001). Qualitative Research and Case Study Applications in
Education. San Francisco: Jossey-Bass Publishers.

Northeast Foundation for Children. (2005). Responsive Classroom. Retrieved
January 8, 2005, flom http://www.responpiveclassroom.corn

Osterman, KR (2000). Students’ need for belonging in the school community.
Review of Educational Research, 70, 32-367.

Pianta, RC. (1994). Patterns of relationships between children and kindergarten
teachers. Journal of School Psychology, 32, 15-31.

Pianta, R.C., Nimetz, S.L., & Bennett, E. (1997). Mother-child, teacher-child
relationships, and school outcomes in preschool and kindergarten. Early Childhood
Research Quarterly, 12, 263-280.

Pianta, R.C. & Walsh, DJ. (1998). Applying the construct of resilience in
schools: Cautions flom a developmental systems perspective. School Psychology
Review, 27, 407-417.

Pierce, T., Baldwin, M.W., & Lydon, J .E. (1997). A relational schema approach
to social support. In G.R. Pierce, B. Lakey, I.G. Sarason, & B.R. Sarason (Eds.),
Sourcebook of Social Support and Personality (pp. 19-47). New York: Plenum Press.

Pierce, G.R., Sarason, B.R., Sarason, I.G., Joseph, H.J., & Henderson, CA.
(1996). Conceptualizing and assessing social support in the context of the family. In
G.R. Pierce, B.R. Sarason, & I.G. Sarason (Eds.), Handbook of Social Support and the
Family (pg. 3-24). New York: Plenum Press.

Phillips, B. (1978). School Stress and Anxiety. New York: Human Sciences
Press.

Richman, J .M., Rosenfeld, L.B., & Bowen, G.L. (1998). Social support for
adolescents at risk of school failure. Social Work, 43, 309-323.

Rosenfeld, L.B., Richman, J.M., & Bowen, G.L. (2000). Social support networks
and social outcomes: The centrality of the teacher. Child and Adolescent Social Work
Journal, 17, 205-226.

Rubin, H.J. & Rubin, LS. (1995). Qualitative Interviewing: The Art of Hearing
Data. Thousand Oaks, CA: Sage Publications.

65

Rutter, M. (1987). Psychosocial resilience and protective mechanisms. American
Journal of Orthopsychiatry, 57, 316-331.

Ryan, RM. (1995). Psychological needs and the facilitation of integrative
processes. Journal of Personality, 63, 397-427.

Ryan, M.R. & Solky, J .A. (1996). What is supportive about social support?: On
the psychological needs for autonomy and relatedness. In G.R. Pierce, B.R. Sarason, &
LG. Sarason (Eds.), Handbook of Social Support and the Family (pg. 249-267). New
York: Plenum Press.

Ryan, R.M. & Grolnick, W.S. (1986). Origins and paws in the classroom: Self-
report and projective assessment of individual differences in children’s perceptions.
Journal of Personality and Social Psychology, 50, 550-558.

Sandler, I.N., Miller, P., Short, J., & Wolchik, SA. (1989). Social support as a
protective factor for children in stress. In D. Belle (Ed.), Children’s Social Networks and
Social Supports (277-307), New York: John Wiley & Sons.

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

Tetzloff, C.E. & Barrera, M. (1987). Divorcing mothers and social support:
Testing the specificity of buffering effects. American Journal of Community Psychology,
1 5, 419-434.

Tietjen, AM. (1994). Supportive interactions in cultural context. In F. Nestrnann
& K. Hurrelmann (Eds.), Social Networks and Social Support in Childhood and
Adolescence (395-407). New York: Walter de Gruyter.

Tigges, BB. (2003). Parental consent and adolescent risk behavior research.
Journal of Nursing Scholarship, 35, 283-287.

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

Yerkes, R.M., & Dodson, J .D. (1908). The relation of strength of stimulus to rapidity of
habit-formation. Journal of Comparative Neurology of Psychology, 18, 459-482.

66

lilliiiljililiiiljliill151116111