Effects of stressors, internal resources, and coping stragegies on the adaptation of families of children with a mental ...

MISSING IMAGE

Material Information

Title:
Effects of stressors, internal resources, and coping stragegies on the adaptation of families of children with a mental handicap
Physical Description:
ix, 130 leaves : ill. ; 29 cm.
Language:
English
Creator:
Paez, Doris, 1964-
Publication Date:

Subjects

Subjects / Keywords:
Parents of children with disabilities   ( lcsh )
Stress (Psychology) -- Management   ( lcsh )
Children with mental disabilities -- Family relationships   ( lcsh )
Children with mental disabilities -- Family relationships   ( fast )
Parents of children with disabilities   ( fast )
Stress (Psychology) -- Management   ( fast )
Genre:
bibliography   ( marcgt )
non-fiction   ( marcgt )

Notes

Thesis:
Thesis (Ph. D.)--University of Florida, 1992.
Bibliography:
Includes bibliographical references (leaves 119-128).
General Note:
Typescript.
General Note:
Vita.
Statement of Responsibility:
by Doris Paez.

Record Information

Source Institution:
University of Florida
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
oclc - 29040655
ocm29040655
System ID:
AA00026341:00001

Full Text












EFFECTS OF STRESSORS, INTERNAL RESOURCES, AND
COPING STRAGEGIES ON THE ADAPTATION OF
FAMILIES OF CHILDREN WITH A MENTAL HANDICAP
















By

DORIS PAEZ


A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA
IN PARTIAL FULFILLMENT OF THE REQUIREMENTS
FOR THE DEGREE OF DOCTOR OF PHILOSOPHY

UNIVERSITY OF FLORIDA


1992


































Copyright 1992

by

Doris Paez














ACKNOWLEDGMENTS

I would like to extend my deepest appreciation to the

many individuals whose encouragement and support contributed

to the completion of this dissertation. Special thanks go

to my chair, Dr. David Miller, who spent much time with me,

offering encouragement, suggestions, and generous assistance

in the analysis and interpretation of the data for this

study. I am extremely grateful to my cochair, Dr. John

Kranzler, whose wisdom and editorial guidance were

invaluable to me. In addition, I would like to thank both

my chairs for their sense of humor and patience.

Words cannot express my gratitude to Dr. Vivian Correa

for sharing with me not only her professional expertise, but

also her home during the past year. I also appreciate Dr.

James Algina and Dr. Ellen Amatea for their professional

feedback and assistance throughout the dissertation process.

My family and friends have also provided me with

extensive encouragement throughout my doctoral program. My

mother, Elsa Paez, and my sister, Diana Paez-Ramos, were a

special source of strength and love. To my father, Pedro

Paez, I offer special thanks for his love and unending

belief in my ability to accomplish anything I wanted. I

also wish to thank my brother-in-law, Leo Ramos, for being a

supportive big brother.


iii








I wish to thank my supporters outside my university

world who often provided me with encouragement, humor, and

diversion from the stress of graduate school: the members

of M.E.C.C.A., Dania Bolt and Emma Debs. I am also grateful

for the support I received from my colleagues in

Hillsborough County. In particular, I wish to thank Mary

Fernandez for her consistent support.

Finally, my husband Scott deserves special thanks for

his love, encouragement, and endurance through a long

process which often involved his personal sacrifice.















TABLE OF CONTENTS

page

ACKNOWLEDGMENTS . . . . . . . . . .. iii

ABSTRACT . . . . . . . . . . .. viii

CHAPTERS

1 INTRODUCTION . . . . . . . . . 1

Context of the Problem ... .................. 1
Factors Related to Adaptation . . . . . 3
Family Stress and Adaptation Models . . . 5
The Double ABCX Model. . . . '.. .. 5
Model of Stress, Adaptation and Ecological
Context . . . . . . . . . 8
Stress-Adaptation Model . . . . . 8
Model of Family Stress ... . ..... 9
Limitations of Research on Family Functioning . 10
Statement of the Purpose . . . . . .. 15
Definition of Terms . . . . . . .. 17
Summary . . . . . . . . . .. 18

2 LITERATURE REVIEW . . . . . . . .. 20

Stress and Adaptation in Families of Children
with a Mental Handicap (MH) . . . .. 20
The Double ABCX Model . . . . . . .. 23
Modifications to the ABCX Model . . .. 24
Empirical Support for the Double ABCX Model 26
Factors Contributing to Stress and Adaptation . 29
Contributions of the Stressor (the Child with
an MH) to Stress or Adaptation (A Factor) 29
Interactions among Stressor Variables . .. 39
Summary . . . . . . . . . .39
Resources and Adaptation (B Factor) ... . 40
Perceptions or Family Definitions of the
Stressful Event (C Factor) . . . .. 45
Resources and Coping Strategies . . .. 47
Limitations of the Double ABCX Model . . .. .48
Modified Versions of the Double ABCX Model . . 50









page

Model of Stress, Adaptation, and Ecological
Context . . . . . . . . .. 50
Stress-Adaptation Model . . . . .. 53
Model of Family Stress . . . . . .. 56
Summary . ..... .................. 58
Family Assessment Instruments ............. 58

3 METHODOLOGY . . . . . . . . .. 62

Data Collection . . . . . . . . 62
Description of the Sample Obtained . . .. 63
Comparison of the Mail-in and Visited
Samples . . . . . . . . .. 63
Subjects . . . . . . . .. 66
Instrumentation . . . . . . . .. 66
Demographic Variables . . . . . .. 67
Family Inventory of Resource Management
(FIRM) . . . . . . . . .. 68
Family Strains Index . . . . o.... . 71
Family Stressor Index... .. . . ... 72
Family Crisis-Oriented Personal Evaluation
Scale (F-COPES) . . . . . . . 73
Family Assessment Device (FAD) . . . .. 77
Data Analysis ... . .............. o.80

4 RESULTS . . . . . . o. . . . 82

Description of Obtained Scale Scores . . .. 82
Factor Intercorrelations . . . . ... o. 87
Stressors and Internal Resources . . .. 89
Stressors and Coping Strategies . . .. 90
Stressors, Internal Resources, Coping
Strategies, and Family Adaptation ... .. 93

5 DISCUSSION . . . . .................. . 95

Discussion of Factor Analytic Results . . .. .95
Relationship between Factors . . . .. 100
Stressors and Internal Resources . . .. 100
Stressors and Coping Strategies .. ... . 103
Stressors, Internal Resources, Coping
Strategies, and Family Adaptation . .. 104
Limitations . . . . . . . ... 108
Future Research . . . . . . . .. 110

APPENDICES

A LETTER TO PRINCIPALS AND TEACHERS . ....... .113

B LETTER TO PARENTS/GUARDIANS . .......... . .114











C SCALE INTERCORRELATION MATRIX . . . . .. .115

D INITIAL FACTOR METHOD: PRINCIPAL COMPONENTS
(UNROTATED) . . . . . . . .. 118

REFERENCES . . . . . . . . . . .. 119

BIOGRAPHICAL SKETCH . . . . . . . . .. 129


vii















Abstract of Dissertation Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Doctor of Philosophy

EFFECTS OF STRESSORS, INTERNAL RESOURCES, AND
COPING STRAGEGIES ON THE ADAPTATION OF
FAMILIES OF CHILDREN WITH A MENTAL HANDICAP

By

Doris Paez

December 1992

Chairperson: Dr. M. David Miller
Cochairperson: Dr. John H. Kranzler
Major Department: Foundations of Education

The purpose of this study was to examine the

functioning of families of children with a mental handicap

as perceived by one parent. The sample was composed of 171

parents from three ethnic groups. Children had one of three

levels of mental handicaps and ranged from 5 to 12 years of

age. The Family Assessment Device (FAD), Family Inventory

of Resource Management (FIRM), Family Crisis-Oriented

Personal Evaluation Scales (F-COPES), Family Stressors Index

(FSTRESS), and the Family Strains Index (STRAINS), as well

as a questionnaire of demographic information, were

completed by the parents.

Results of this assessment battery were subjected to

factor analysis. Three factors emerged, reflecting Internal

Resources, Coping Strategies, and Family Adaptation.


viii








Multiple regression analyses were then conducted to examine

the relationship between family stressors and the three

factors. The results revealed a significant relationship

between Internal Resources, a child's age, parent's age, and

ethnic group. Coping Strategies was found to be only

related to child's severity of mental handicap. Moreover,

ethnicity was found to be significantly related to Family

Adaptation. Child's gender was not found to be

significantly related to family functioning. In addition,

the mediating role between stressors and family adaptation

of Internal Resources was supported in this study. However,

the same role was not supported for Coping Strategies.

These results have implications for theory and practice,

particularly for those professionals involved in family-

focused interventions.

















CHAPTER 1
INTRODUCTION

Context of the Problem

The focus and scope of intervention services for

children with a mental handicap (MH) have recently expanded

to include the families of those children. This expansion

is the result of an awareness on the part of researchers,

school personnel, and policy-makers that a child with an MH

is part of a larger family system in which the child both

influences and is influenced by the family. Thus, the

success of any intervention is related integrally to the

family. Many authors have noted that even the best

interventions sometimes have little impact because the

family was not considered (e.g., Bailey & Simeonsson, 1988;

Dunst, Vance, & Cooper, 1986). The lack of consideration of

the family is largely due to the limited information that is

available concerning the functioning of families of children

with an MH. Such information is needed to guide school

personnel in targeting specific aspects of family

functioning in order to maximize the effectiveness of

intervention strategies (Gallagher & Bristol, 1989).

In fact, family functioning is such an important

component in the delivery of services to young

children with an MH that policies regarding evaluation

1








procedures and provision of family services have recently

been created. For example, the Education of the Handicapped

Amendments Act of 1986 (P.L. 99-457, Part H) mandates that

the strengths and needs of families of handicapped infants

and toddlers be identified and included in the educational

plans for these youngsters. Although this law is geared

toward use with the families of handicapped infants and

toddlers, some believe that it will soon be extended to

include families of elementary and secondary students with

handicaps (e.g., Darling & Seligman, 1989; Turnbull, 1990).

Researchers and service providers in special education

have looked to such fields as sociology or marriage and

family therapy for guidance in conceptualizing and assessing

the functioning of families of children with handicaps

(Sheehan, 1989). Results of research from these fields show

that such families experience considerable stress (Minnes,

1988a, 1988b).

In addition, family functioning is often described in

terms of the relationship between stress and adaptation

(Minnes, 1988b). Research on the family has focused

traditionally on the poor or unhealthy adaptation of

families to the challenges and demands of life with an MH

child (Dunst, Trivette, & Deal, 1988). In recent years,

however, researchers have become interested in factors

related to successful adaption of these families (Bristol,

1984, 1987; Gallagher & Bristol, 1989; Gallagher, 1988;

Kazack & Marvin, 1984). Investigators have found that many








families report better adaption as a result of having a

member with a handicap (Dunst et al., 1988; Summers, Behr, &

Turnbull, 1989; Turnbull & Turnbull, 1990). Such findings

have led to increased interest in the assessment of factors

related to the adaptation of families of children with an MH

(Gallagher & Bristol, 1989).

Factors Related to Adaptation

A wide variety of factors have been found to

contribute to the adaptation and stress experienced by

families of children with an MH (see Minnes, 1988b, for a

review). These factors include child and parent

characteristics, family resources, and family coping

strategies, among others.,

Mixed results have been found among studies of the

relationship between family stress and characteristics of

the handicapped child, such as the child's age (Bristol,

1979; Wikler, 1986a), gender (Gallagher & Bristol, 1989),

severity of disability (Minnes, 1984, 1988b), and behavioral

problems (Friedrich, Wilturner, & Cohen, 1985; Orr, Cameron,

& Day, 1991). For example, some researchers have reported

that families with male handicapped children tend to

experience more stress than families with female handicapped

children (Bristol, 1984; Farber, 1959; Roesel & Lawlis,

1983). However, Friedrich (1979) found the opposite results

(i.e., more stress in families of female children) and

Beckman (1983) failed to substantiate the relationship

between gender and family stress.








Several parent characteristics have been linked to

differences in the family stress reported, namely, age of

the parent (Bailey & Simeonsson, 1988; Dunst et al., 1988;

Minnes, 1988a; Turnbull & Turnbull, 1990), ethnicity

(Friedrich et al., 1985), general family coping strategies

(Rabking & Steunings, 1976), perceptions of the

developmental course and demands of the child's disability

(Petersen, 1981), and quality of marital relationship

(Abbott & Meredith, 1986; Friedrich & Friedrich, 1981;

Friedrich et al., 1985; Petersen, 1981).

Differences in reported family stress have also been

found to vary according to the family's socioeconomic status

(SES), perceived financial well-being, and social class

(Minnes, 1988b). However, discrepant findings are evident

in this area. For example, significant relationships were

not found between any parent or family variables (e.g., age,

income) and parental mental and physical well-being (Dunst,

Trivette, & Leet, 1986).

In addition to these findings, strong relationships

have been established between family adaptation and the

types (internal or external) and accessibility of family

resources and coping strategies (Abbott & Meredith, 1986;

Friedrich et al., 1985; Mink, Nihira, & Meyers, 1983, 1988;

Minnes, 1984; Nihira, Meyers, & Mink, 1980; Peterson, 1981).

For example, Abbott and Meredith (1986) found that

adjustment to stress in families of children with an MH is

strongly related to the parents' self-esteem, strength of








marital relationship, and ability to reframe stressors in

positive ways.

The discrepant findings across these studies are

typical of family research. Many researchers attribute such

discrepancies to methodological differences (Beckman, 1983;

Minnes, 1988b). Others argue that differences in family

adaptation are related to family resources and coping

strategies (Dunst, Trivette & Deal, 1988; McCubbin &

Patterson, 1982; Minnes, 1988b; Turnbull & Turnbull, 1990).

Because these relationships are so complex, models of family

stress and adaptation have been developed.

Family Stress and Adaptation Models

Several models of family stress and adaptation have

been specifically proposed for the study of families of

children with handicaps. The most prominent model is the

Double ABCX model (McCubbin & Patterson, 1982, 1983). The

other models that have been proposed can be seen as modified

versions of this model (Crnic, Friedrich, & Greenberg, 1983;

Farran, Metzger, & Sparling, 1986; Singer & Irvin, 1989).

These models are discussed in the next section.

The Double ABCX Model

The Double ABCX, or Family Adjustment and Adaptation

Response Model (FAAR), proposed by McCubbin and Patterson

(1983), is an expansion of Hill's (1949, 1958) classic

theory of family stress. According to Hill, the

characteristics of the stressor event (A), the family's

internal resources for meeting a crisis (B), and the








family's definition of the stressor event (C), all

contribute to the prevention or precipitation of family

crisis (X). McCubbin and Patterson (1982, 1983) extended

the ABCX Model by adding factors to address additional life

stressors and strains that shape the course of adaptation

(aA), such as the social and psychological resources

families acquire and the coping strategies which they employ

(bB), the meaning families attach to stressors and strains

(cC), and the possibility of both positive and negative

family outcomes (xX).

Gallagher and Bristol (1989) stated that the Double

ABCX model is an adequate conceptual framework for studying

families of children with handicaps. First, the development

of children with handicaps and their families over time is

recognized in this model which is consistent with the

special education literature. Second, this framework

emphasizes two additional factors, active and passive

coping, which reflect the actions taken by families in

response to stressors or the acceptance of stressors by

families. Third, and finally, the possibility that healthy

adaptation, rather than pathology, may characterize the

family's response to stressors is addressed. McCubbin and

his associates have created measures of each of the factors

in the model, many of which have demonstrated validity and

reliability. Furthermore, the Double ABCX model has

accounted for the results found in several empirical studies

of families with handicapped children (Bristol, 1984, 1987,









1989; Comeau, 1985; McCubbin & Huang, 1989; Orr, Cameron, &

Day, 1991; Petersen, 1981).

Despite the proven applicability of the Double ABCX

model, problems in its conceptualization and use are

evident, specifically concerning the parameters of

resources, coping skills, and perceptions. While some

conceptualized these factors as McCubbin and his associates

did (Orr, Cameron, & Day, 1991), others propose that these

factors overlap (Dunst et al., 1988; Gartner, Lipskey &

Turnbull, 1991; Mitchell, 1991; Singer & Irvin, 1989). For

example, Dunst et al. (1988) have combined internal

resources and coping skills under one factor, referred to as

family functioning style, which is separate from external

resources or social supports.

In addition, discrepancies exist in the distinction

between stressor and adaptation. For example, level of

stress has been reported as an indicator of the stressor

event/pile-up (aA) and as a measure of adaptation (x)

(Wikler, 1986b). Furthermore, the interaction of factors in

the Double ABCX and the ecological context is unclear. Some

suggest that the Double ABCX does consider ecological

context (Gallagher & Bristol, 1989), while others state that

it does not (Crnic et al., 1983).

To address the limitations of the Double ABCX model,

modified versions of the Double ABCX have been proposed for

studying families of mentally handicapped children (Crnic et

al., 1983; Farran, Metzger, & Sparling, 1986; Singer &

Irvin, 1989). In these models, the factors of the Double








ABCX model were altered to highlight different aspects of

family functioning.

Model of Stress, Adaptation and Ecological Context

Crnic, Friedrich, and Greenberg (1983) proposed the

Model of Stress, Adaptation and Ecological Context. Drawing

from the literature in the areas of stress, individual

coping, and ecological influences on development and

functioning, Crnic et al. added the concept of ecological

context to the Double ABCX model. Ecological context refers

to the conditions and settings in which the family exists

(e.g., home, school, etc.). In addition, coping strategies

and resources (internal and external) are conceptualized in

this model as one factor, termed coping resources. By

collapsing these two concepts, however, the authors failed

to distinguish between family resources and coping

strategies. For example, a family may have a large extended

family (external resource), but may not ask them for

assistance (coping strategy). Furthermore, the individual

contributions of resources and coping strategies to family

adaptation, as well as the interaction between them, become

difficult to delineate in this model. Another model

proposed for the study of families of children with an MH is

the Stress-Adaptation Model.

Stress-Adaptation Model

Farran, Metzger, and Sparling (1986) proposed the

Stress-Adaptation Model which is another modified version of

the Double ABCX model. In the Stress-Adaptation Model,

however, the basic factors of the Double ABCX are redefined








and relabeled. The Stress-Adaptation Model is comprised of

three factors: (a) Demands stressorr event and changes

associated with the event), (b) Adaptive Capacity (coping

experiences, definition of the stressor event, and social

support), and (c) Adaptation (functioning of the

parents/family and child). The factors in this model are

held to be sequentially related, which excludes possible

interactions. In addition, this model addresses separate

child and parent outcomes. Separating these outcomes

highlights that individual family members are being

addressed rather than whole family outcomes. Another

limitation of this model is that some overlap is evident

between the Demand factor and the Adaptive Capacity factor.

For example, the personal evaluation of self and future,

which is subsumed under the Demands factor, appears to be

conceptually similar to the "personal protective factors,"

which are incorporated into the Adaptive Capacity factor.

Model of Family Stress

More recently, Singer and Irvin (1989) modified the

Double ABCX family model to make it parallel to the model of

stress and adaptation in individuals presented by Billings

and Moos (1984). Included in the Model of Family Stress is

the concept of adaptation to individual stress proposed by

Folkman and Lazarus (1984). As in the Double ABCX model,

family adaptation is a function of the intra-family and

community resources and of the family's appraisal and coping

response.






10

Several departures from the Double ABCX model, however,

are evident. First, resources have been separated into

internal and external, with both being separate from coping

response. Although not labeled as such, there are,

therefore, two separate "B" factors: intra-family resources

and community resources. Thus, it may be useful to label

one factor "Bl" and the other "B2." Second, the family

stressors are defined as problems directly related to the

child's specific handicap (e.g., behavior problems, ill

health, appearance) or the family (e.g., social isolation,

money worries) or both. Finally, the model does not contain

an A to B to C to X path, as proposed in the ABCX and Double

ABCX models, but, rather, a B to A to C to X path. In this

model, the intra-family (Bl) and community resources (B2)

simultaneously define the stressor (A). The stressor (A),

in turn, impacts family appraisal and coping responses (C)

resulting in positive or negative adaptation (X). Although

the model is described sequentially, Singer and Irvin (1989)

stated that each factor could affect every other factor.

For example, intra-family resources might influence not only

the stressor (A), but also the community resources (B2), the

family appraisal and coping response (C), and adaptation

(X).

Limitations of Research on Family Functioning

At the root of the conceptual and methodological

problems that plague research on families of children with

an MH is the fact that "the concept of family itself is too

rich, complex, and abstract to be captured by any one






11

definition, instrument or method" (deGruy & Dickinson, 1991,

p. 17). Consequently, the definition of a "family" varies.

Such differences are to be expected and are considered by

some as "both necessary and good" (deGruy & Dickinson,

1991). Moreover, differences in definitions have resulted

in sampling differences, which limit the generalizability of

results of studies in this area.

The weaknesses that have plagued research on families

with children with an MH are also related to difficult

pragmatic issues (Stoneman & Brody, 1990). Some of these

obstacles include difficulty recruiting families, due to the

low incidence of these children in the general population,

and the heterogeneity among children classified as MH.

These problems are further compounded by small sample sizes,

ill-defined populations, grouping of MH children with

children with a diversity of handicapping conditions, and

heterogenous family situations. Furthermore, studies of

families of children with an MH have included broad age

ranges of children (0 to 21 years), a focus on only one

diagnostic category (e.g., trainable mentally handicapped),

different respondents (mothers, fathers, grandmothers), and

various children's educational placements (ranging from

self-contained classrooms to full mainstreaming).

Another source of complexity arises from the diversity

of theories of family functioning. As observed in the

various family stress and adaptation models, each model

posits a somewhat different set of functions or patterns of

family operations. Because the family variables assessed






12

inevitably follow the model used by a researcher, results of

studies in this area are difficult to compare. For example,

many researchers have examined the multiple components of

the functioning of families of children with an MH (e.g.,

Bristol, 1987; Flynt & Wood, 1989; Minnes, 1988b; Orr,

Cameron, & Day, 1991; Wikler, 1986a, 1986b). However, these

studies differ considerably with regard to the definition of

family stressors, resources, coping strategies and

adaptation employed. The definitions vary as a result of

differences in the parameters of the specific model employed

(i.e., using the Double ABCX model, as opposed to the

Stress-Adaptation model) or according to differences in

interpretations of one particular model (e.g., Orr, Cameron,

& Day, 1991; Wikler, 1986b).

Another level of complexity that arises is related to

the assessment of family functioning. Typically, the

functioning of families of children with an MH is assessed

through self-report measures. Self-report instruments are

used because they are time- and cost-efficient. However,

certain aspects of family functioning may be best sampled by

other means, such as direct observations and interviews

(Gartner, Lipsky, & Turnbull, 1991; Minnes, 1988a).

Moreover, forced-choice measures, which are common in this

research, limit the nature of the information that can be

obtained. This problem is further compounded by the fact

that few measures with adequate reliability are available

for assessing family functioning (Wikler, 1986b).






13

Variability in the measures used to assess resources in

studies of families of children with an MH is also evident.

Some widely used self-report scales lack validity or

reliability data (e.g., Petersen, 1981; Wikler, 1986b).

Other instruments, however, are grounded in theory and have

demonstrated validity and reliability (e.g., McCubbin,

Comeau, & Harkin, 1981; McCubbin, Larson & Olson, 1982).

Part of the problem is that the validity of self-report

instruments is frequently limited by social desirability,

which refers to a bias on the part of respondents to answer

questions in a manner they believe is socially acceptable.

Researchers in this area also typically rely on mothers as

respondents, mainly because fathers are often reluctant to

participate (Stoneman & Brody, 1990). Research relying upon

maternal perceptions is important, however, as mothers are

typically the primary caregivers and most involved in

intervention efforts (Dunst, Trivette, & Leet, 1986;

Stoneman & Brody, 1990).

Several authors have suggested that more studies are

needed in which family functioning models are

operationalized using validated self-report instruments

(Gallagher & Bristol, 1989; Minnes, 1988a; Wikler, 1986b).

Other authors believe that family functioning should not be

assessed by self-report scales, but through interviews and

informal observations (Turnbull & Turnbull, 1990). Yet

others believe that a combination of self-report scales,

interviews, and observations should be used in assessments








of family functioning (Stoneman & Brody, 1990). Still

others believe that "just as it is not necessary to select a

single definition of family that must serve all purposes,

neither is it necessary to select a single method as the

best for collecting family data" (deGruy & Dickinson, 1991,

p. 18). Furthermore, published instruments are more

valuable when they have been used in a variety of settings

and for a variety of populations.

Beyond the conceptual and methodological problems

encountered in studies of families of children with an MH,

issues specific to the public school setting must be

considered. First, given shortages of personnel, public

schools need time- and cost-efficient objective methods for

assessing family functioning. Second, most researchers who

study families with handicapped children have gathered their

data in clinic- or university-based classroom settings or

both, which renders their results less salient to the public

school setting. Consequently, a need for information

regarding the families served through public schools exists.

Moreover, within the public schools a variety of children

with an MH (i.e., who differ in severity of mental handicap)

are served in a variety of classroom settings (e.g., self-

contained versus mainstreaming programs). Distinctions

between these children should be made because such

differences may account for differences in family

functioning (Crnic et al., 1983).









Statement of the Purpose

The purpose of this study was to further our

understanding of adaptation of families with children with

an MH. While several modifications of the Double ABCX have

been proposed by contemporary family researchers (Crnic,

Friedrich, & Greenberg, 1983; Farran, Metzger, & Sparling,

1986; Singer & Irvin, 1989), this study was an investigation

of the relationships among the factors of various models.

Of particular concern was the identification of the factors

underlying the particular family assessment instruments with

this population. Consequently, the factors were

operationalized differently than in previous studies (see

Figure 1).



^ INTERNAL RESOURCES



STRESSOR

Child's Age FAMILY
Child's Gender ADAPTATION
Ethnicity
Parent's Age
Sources of Financial Support
Severity of Mental Handicap /
(Educable, Trainable,
Severe/Profound)
General Family Stress



COPING
STRATEGIES


Figure 1. Operationalized Model.








In addition, this study addressed some of the

methodological problems evident in family studies by using a

different sampling procedure. The sample here was composed

of families of children between 5 and 13 years of age with a

mental handicap receiving instruction in self-contained

classrooms. Thus, the effects of age, severity of handicap,

and the ecological context were addressed. The possible

effects of ethnicity on family adaptation and buffering

factors were addressed by including Caucasian, Hispanic, and

African-American families in the sample. Furthermore, given

that parental age has been found to be an important marker

variable, the effect of the caregiver's age was studied.

Finally, because the public schools ultimately will be

legally mandated by P.L. 99-457 to assess and provide

services to the families of mentally handicapped children,

the families were recruited through the public schools.

Demographic information and responses on three

published self-report family assessment instruments were

collected to address the following research questions:

1. Is there a significant relationship between

stressor variables--i.e., demographic variables (child's

age, child's gender, severity of MH, ethnicity, parent's

age, and sources of financial support) and general family

stress--and internal resources?

2. Is there a significant relationship between

stressor variables and coping strategies?








3. Is there a significant relationship between

stressor variables, internal resources, and coping

strategies and family adaptation?

Definition of Terms

The following definitions are given to define and

clarify terms as they are used in this study.

A family is comprised of the persons living in the same

households the child with an MH. For the purposes of this

study, the results regarding the family are based on the

perception of the parent or guardian who responded to the

questionnaires.

Children with a mental handicap (MH) have Intelligence

Quotients (I.Q.) which are two standard deviations below the

mean on standardized instruments. Mental Handicap is also

referred to as mental retardation in the literature. Three

severity levels of MH are distinguished in this study:

Educable or Mild (I.Q. range, 55 70), Trainable or

Moderate (I.Q. range 35 54), and Severe/Profound (I.Q.

<35).

Stressors refer to the presence of a child with an MH

and general life stressors faced by families in the past

year (e.g., divorce) (Minnes, 1988b).

Internal Resources are intra-family characteristics or

qualities that can be mobilized to deal with stressors

including self-esteem, sense of competence, and sense of

financial well-being (Dunst et al., 1988; Singer and Irvin,

1989; Turnbull & Turnbull, 1990).








Coping Strategies are overt and covert behaviors in

response to a stressor (e.g, reframing ability, acquiring

and maintaining social supports, seeking spiritual supports)

(McCubbin & Thompson, 1987; Singer & Irvin; 1989).

Family Adaptation refers to the structural and

organizational properties of the family and transactions

among family members which have been found to distinguish

between healthy and unhealthy families (e.g., role

differentiation, communication, affective responses)

(McCubbin & Thompson, 1987).

Summary

The important role of the family in child-focused

interventions recently has been established. Furthermore,

legislation has been passed mandating the assessment of

family functioning and the inclusion of family-focused

interventions in the special education programs of young

children with handicaps. Consequently, researchers and

service providers have become interested in the functioning

of families of children with an MH.

Family stress and adaptation models are useful for

examining family functioning. In particular, these models

delineate the relationship between stress and adaptation to

such stress, as well as the role of factors that mediate

stress. However, the models and research studies, upon

which they are based, are plagued by conceptual and

methodological problems. There is, therefore, a need for

further examination of the relationship between stress and

adaptation in families of children with an MH. The purpose

of this study is to investigate the relationship among







19

internal resources, coping strategies, and adaptation of

families of children with an MH. The results of this study

may have direct implications for school personnel involved

with family-focused interventions for families of children

with an MH.

A review of the literature relevant to this study is

presented in Chapter 2. In Chapter 3 the methodology is

discussed. Results and implications of this study are

presented in Chapters 4 and 5.














CHAPTER 2
LITERATURE REVIEW

The purpose of this chapter is to review the literature

related to adaptation in families of children with a mental

handicap (MH). The first section presents a discussion of

the relationship between stress and adaptation in families

of children with an MH. Next, the Double ABCX Model, the

most prominent model used to examine family stress and

adaptation in families of children with an MH, is reviewed.

In the third section, the Double ABCX Model is used as a

framework for examining past research findings concerning

factors contributing to stress and adaptation of families of

children with an MH. The fourth section is a critique of

the Double ABCX model and the studies in which it has been

used. In the fifth section, modified versions of the Double

ABCX model are examined. In the sixth and final section,

the strengths and weaknesses of the most prominent family

assessment instruments are discussed.

Stress and Adaptation in Families of Children with a Mental
Handicap (MH)

The most commonly described effect of having a child

with an MH is the stress experienced by the family (see

Hanson & Hanline, 1990, for a review). However, like the

study of individual stress, the study of stress in families

of children with an MH has been plagued by definitional






21

problems (Bailey & Simeonsson, 1988; Minnes, 1988b). Stress

has been viewed both as a function of the event itself

(i.e., having a mentally handicapped family member) and as a

the family's response to the event. Moreover, hardships

associated with the event (e.g., financial problems) are

frequently subsumed under the concept of family stress

(McCubbin, Cauble, & Patterson, 1982).

In addition to the problems of definition,

methodological problems (e.g., sampling, instrumentation,

and data collection procedures) make it difficult to

ascertain how families of children with an MH are different

from other types of families (e.g., families of children

with other handicaps and families of children without

handicaps) with regard to stress. When stress data of

families of children with an MH are pooled with that of

families with children with other handicaps, the empirical

evidence regarding the stress of families of children with

an MH has been inconsistent. Some investigators have found

that families of children with an MH have heightened

parental stress (e.g., Dyson & Fewell, 1986; Friedrich &

Friedrich, 1981; Kazak & Marvin, 1984; McKinney & Peterson,

1987; Minnes, 1984, 1986; Waisbren, 1980) and deteriorated

family relationships (Farber 1959, Friedrich & Friedrich,

1981; Gath, 1977). Others have reported that these families

do not experience greater stress (Frey, Greenberg, & Fewell,

1989; Gown, Johnson-Martin, Goldman, & Appelbaum, 1989;

Harris & McHale, 1989; Salisbury, 1987) or less marital






22

satisfaction (Kazak, 1987; Kazak & Marvin, 1984). Hence, no

definitive statement regarding the stress experienced by

families of children with an MH can be made. Therefore,

more studies are needed before a substantive conclusion can

be reached.

Only few researchers have contrasted families of

children with an MH with other types of families. Results

generally indicate that families of children with an MH

experience more stress than families of children with other

primary handicapping conditions (e.g., physical) or families

with children of normal intelligence (Bristol, 1989; Minnes,

1988a).

Evidence also exists suggesting that many families of

children with an MH adapt in healthy and functional ways

(Kazak & Marvin, 1984; Turnbull, 1990; Turnbull & Turnbull,

1986). Furthermore, many families report specific positive

contributions from having children with handicaps (Abbott &

Meredith, 1986; Summers, Behr, & Turnbull, 1989; Turnbull &

Turnbull, 1986, 1990). For example, many families report

increased family cohesiveness (Abbott & Meredith, 1986).

Thus, mixed results regarding the stress experienced by

families of children with an MH have been obtained.

Families of children with an MH have been described as

experiencing more stress, as being less capable of adapting

to stress, and as adapting in healthy ways. These

discrepant results appear to be a function of definitional

and methodological differences across studies of stress and








adaptation in families of children with an MH. Therefore,

there is a need for models which can explain the differences

in results and guide future research.

The Double ABCX Model

The most prominent model used to explain the adaptation

of families of children with an MH is the Double ABCX or

Family Adjustment and Adaptation (FAAR) model (McCubbin &

Patterson, 1982, 1983). The foundation of the Double ABCX

is based on Hill's classic ABCX Model (1949, 1958) which was

derived from studies of war-induced familial separations.

Hill (1949) dismissed the notion that a direct relationship

between stress and adaptation existed and, instead, proposed

that two major mediating variables (resources and

perceptions) protect the family from becoming dysfunctional

and possibly facilitate positive adaptation. Thus, Hill's

ABCX Model was composed of the stressor or stressor event

(A), the family's resources for meeting crisis (B), and the

family's definition of the stressor (C), each of which

contribute to the prevention or precipitation of family

crisis, (X). The stressor event (A) includes (a) the source

of the event (within or outside the family), (b) the effects

of the event upon the configuration of the family (the

addition or loss of a family member), and (c) the type of

event (normative/expected or nonnormative/unexpected). The

B factor was conceptualized by Hill (1958) as "the adequacy-

inadequacy of family organization" (p. 29). In his

synthesis of the family resources literature, Burr (1973)






24

described the B factor as the family's ability to prevent a

change in the family system from becoming a crisis or

disruption to the family system. The C factor was defined

by Hill (1949) as the subjective meaning the family added to

the stressor event. That is, the degree to which the

situation is seen by a family as a threat to the family

system. The X factor, crisis, was viewed as any sharp or

decisive event for which current family patterns are

inadequate, the amount of disruption in the family system,

and the extensiveness of the impact of the stressor on the

family's functioning (Hill, 1949).

Modifications to the ABCX Model

Despite the usefulness of the ABCX model, it did not

account for the evolving nature of families, the effects of

the accumulation of stressor and strains over time, and the

build-up of family resources. Recognizing the need for a

more "dynamic" model, McCubbin and Patterson (1982, 1983)

developed the Double ABCX model (see Figure 2), which

incorporates many of the key components of the earlier

model.

McCubbin and Patterson (1982, 1983) focused on the

family 's efforts to recover from crisis or stressful

situations. The Double ABCX model added the dimensions of

additional life stressors and strains, which are termed

pile-up of demands (aA), the resources the family acquires

and the family's coping strategies (bB), including the

meaning families attach to stressors and strains (cC), and









the family's adaptation along a continuum from negative to

positive (xX).


Figure 2. The Double ABCX Model (McCubbin & Patterson,
1982, 1983).


In addition to focusing on the evolving nature of the

child and the family, several other features of the Double

ABCX model make it uniquely suited for studying families of

children with an MH. First, the assessment of active and

passive coping strategies is possible. Second, families can

be characterized by healthy adaptation, rather than

unhealthy adaptation (Dunst et al., 1988). Finally, valid






26

and reliable instruments have been created specifically for

assessing the factors in this model.

Empirical Support for the Double ABCX Model

The majority of the empirical support for the Double

ABCX model comes from studies with samples of families of

chronically ill children (e.g., Comeau, 1985; McCubbin,

1986; McCubbin & Huang, 1989; McCubbin, 1989) or other

handicapping conditions, including mental handicaps (e.g.,

Bristol, 1987; Gallagher & Bristol, 1989; Petersen, 1981).

Few researchers have focused solely on families of children

with an MH (Flynt & Wood, 1989; Orr et al., 1991; Wikler,

1986a, 1986b). Wikler (1986a), for example, applied the

Double ABCX model to families with children with an MH. The

dependent variable, measures of factor X, in Wikler's study

was family stress. The stressor was conceptualized as the

child's mental handicap and transition stage (e.g., onset of

puberty, onset of adulthood), which was measured by the

child's age. Family stressors unrelated to the child's

mental handicap were included to measure the Aa factor. For

the B and C factors, Bristol created a 20-item checklist,

with 10 items for family resources and 10 for family

perceptions. She found that not only the B and C factors

significantly related to family stress (X), but they

explained more variation than did the child's age. However,

Wikler also found that families at transition points

reported higher levels of stress than families not at

transitions points. For example, families with children








with an MH experiencing the onset of puberty reported having

more stress than parents of elementary age children with an

MH. Thus, Wikler's results support the relationship between

the stressor (A) and stress (X) and the buffering and

significant effect of B and C on X.

While Wikler (1986a) addressed each factor in the

model, several conceptual and methodological problems were

evident. First, although the outcome measure was presented

as a general family functioning index, it actually focused

on two levels: one individual (i.e., maternal stress) and

two individuals (i.e., parent-child interaction), rather

than the whole family unit. Second, it is possible that

those life stressors (e.g., unemployment, divorce, death of

family member) and strains (e.g., financial worries) in the

aA or general A factor may be better conceptualized as

resources (or the lack thereof). This highlights the

problem of overlap among the concepts in the model (Gartner,

Lipsky, & Turnbull, 1991). Third, several items sample both

the resources and perception domains. Another significant

problem is that the checklist used does not have proven

validity or reliability data. Fourth, while the effect of

the age of the child was considered, differences due to the

severity of the handicap were not. Finally, as with most

studies in this area, only mothers completed the

questionnaire. Thus, only one individual's view of the

family was obtained. Some researchers suggest that

gathering information from more than one family member is

necessary because mothers, fathers, and siblings of children









with an MH have been found to differ on occasion in their

perceptions of family functioning (see Turnbull & Turnbull,

1990).

In spite of these criticisms, Wikler's study does

illustrate how family stress theory can predict some kinds

of family stress and adjustment. Other investigators have

also supported the use of the Double ABCX framework for

families with children with an MH (Bristol, 1987; Orr,

Cameron, & Day, 1991; Petersen, 1981). For example, the

predictive power of the Double ABCX model was also

demonstrated by Orr et al. (1991) in a study of 86 families

of children with a moderate MH. The focus of this study was

to evaluate the paths between factors as proposed in the

Double ABCX model. Factor A was defined as frequency of the
P
child's behavioral problems and Factor B as the social-

psychological resources of a family. Factor C was

operationalized as the family's capability to redefine a

stressful situation, as measured by a subscale of a coping

measure. Finally, Factor X was defined as the degree of

stress in the parent-child relationship. Using path

analysis, results revealed significant relationships between

all the factors of the Double ABCX model. Interestingly,

however, the factors followed an ACBX path, as opposed to an

ABCX path. Orr et al. argued that from an intervention

perspective these results seem reasonable because the

effectiveness of resources in reducing stress is dependent

upon how the family/parent define and interpret the child's

needs, level of functioning, and problem behaviors.








Therefore, the Double ABC may need to be reconceptualized or

modified.

A comparison of the Wikler (1986a) and Orr et al.

(1991) studies highlights additional problems common in

family research. First, the operationalization of the

Double ABCX model varies among studies. For example, the

stressor in the Orr et al. study was behavior problems

(alterable characteristic), as opposed to Wikler's

definition of the stressor as the age of the child

(unalterable characteristic). Furthermore, in some studies

(e.g., Orr et al., 1991) the C factor is considered to be

only one type of family perception (refraining) and excludes

other facets of perception (e.g., passive appraisal) and

coping mechanisms (e.g, problem-solving skills). Several

methodological limitations also stem from the samples

employed, including the broad age ranges of the children (5

to 21 years), the focus on only one diagnostic category

(trainable mentally handicapped), the variety of respondents

(mothers, fathers, grandmothers), and differences in the

children's educational placements (ranging from self-

contained classrooms to full mainstreaming). These

limitations will be further examined in another section of

this review.

Factors Contributing to Stress and Adaptation

Contributions of the Stressor (the Child with an MH) to
Stress or Adaptation (A Factor)

Although universal agreement on this topic has not been

reached (Bristol & Gallagher, 1989), it appears that family

stress or adaptation varies as a function of several family









variables. These variables include child and parent

characteristics, such as the severity of the child's mental

handicap (e.g., educable, trainable, or severe/profound) and

child and parent gender, as well as the family's life cycle

stage, economic status, and ethnicity (Beckman, 1983;

Minnes, 1988b).

Severity of handicap. Minnes (1984, 1986) found that

severity of handicap was significantly related to the degree

of stress reported by parents. In contrast to families of

children with a moderate or severe MH, parents of children

with a mild MH reported significantly fewer limits on family

opportunities. Parents of children with a mild MH also

report less concern regarding life-span care and terminal

illness stress than parents of children with a severe MH.

In addition, parents of children with a mild or a moderate

MH reported significantly less stress associated with

physical limitations than parents of children with a severe

MH.

However, an inverse relationship between degree of

handicap and parental stress was also reported (Bristol,

1984; Weller, Costeff, Cohen, & Rahmar, 1974). Parents of

children with a more severe MH are less likely to deny their

child's problems and, as a consequence, experience less

stress than parents of children with a milder MH, who may

hold unrealistic expectations due to their relative

diagnostic uncertainty and a more normal appearance. These

studies point to the need for more research that








systematically controls for the degree of handicap. It

should be noted that Minnes (1988b) also suggested that, in

conjunction with degree of handicap, parental perception of

the handicap should be considered because it may influence

the degree of stress. For example, a parent of a child with

a severe MH with the ability to positively reframe stressful

situations may report less stress than a parent of a child

with a mild MH lacking this ability. The important

contribution of the family definitions of the stressor event

will be discussed later in the perceptions section.

Child's gender. Research on the relationship between

parental stress and the gender of a child with an MH are

unclear to date (Minnes, 1988b). Early work in this area

indicated that couples with a male child with an MH

experienced more marital problems than those with a female

child with an MH (e.g., Farber, 1959). Roesel and Lawlis

(1983) found that young parents with a first born male with

a genetic and mental handicap were at higher risk for

divorce than parents of female children with similar

handicaps. Bristol (1979) found lower integration (an

internal resource) in families with a male autistic child

than with a female autistic child. In contrast, Friedrich

(1979) found that mothers of female children with an MH

experienced more stress than mothers of male children with

an MH, while Beckman (1983) also did not find a relationship

between a child's gender and mother's stress. The apparent

discrepancy across studies may be a function of differences








in the ages of the children in the samples, however, as

noted by Beckman (1983). Children in the Farber and Bristol

studies represented a broad age range (4 to 21 years), while

children in the Beckman sample ranged from 6 1/2 months to 3

1/2 years. Beckman suggests that the child's gender may be

less important to parents than degree of handicap in the

early stages of development.

Unlike Beckman (1983) and Friedrich (1979), Bristol

controlled for the level of child dependency and found that

the greater stress reported by parents of male children also

may reflect the requirement for greater management that

often accompanies these children as they mature.

Nevertheless, it is difficult to explain Friedrich's (1979)

findings. As with severity of handicap, the differences in

parental stress may be more related to parent's perception

of the child's handicap than to gender (Minnes, 1988b).

Parent's gender. Few researchers have examined

differences between mother's and father's reported stress

and adaptation in studies of families of children with an MH

(e.g., Bristol & Gallagher, 1986; Minnes, 1988a). This

situation exists for several reasons. First, most

researchers have employed samples comprised mainly of

mothers (Dunst et al., 1988). Several researchers (Bristol

& Gallagher, 1989; Stoneman & Brody, 1990) have stated that

despite heightened awareness of the importance of father's

roles in the lives of children with an MH, fathers tend not

to be included because it is difficult to gain access to








them. When fathers have been included, the differential

aspects of mother versus father responding was not

considered (e.g., Orr et al., 1991). Instead, the focus of

these studies has been on the differences between fathers of

children with an MH and fathers of children with normal

intelligence.

Life-cycle stage. Minnes (1988b) stated that in

addition to focusing on the intellectual level and gender of

the child, research on stress in families must consider the

family's life-cycle stage. Differences among families in

terms of their responses to the child with an MH exist

across the life-cycle stage of the child and the family

(Gartner, Lipsky, & Turnbull, 1991). Turnbull and Turnbull

(1990) have related life-cycle stages to the common concerns

that relatives have about children with an MH. In effect,

there are two intertwined and interacting life-cycles--that

of the family unit and that of individual members, including

the child with an MH.

Life-cycle stage can be defined by the age of the child

and the age of the parent. Although definitions of family

stages differ, in general, the life cycle stages for

families of children with an MH (when defined by the child's

age) are preschool years (3-5 years), entering school years,

(6-11), adolescence (12-18) and adulthood (19-21) (Beavers,

Hampson, Hulgus, & Beavers, 1986). It should be noted that

because the life-cycle stages of these families are entwined

with educational time lines, the actual ages included in








each stage could vary with differences in criteria from

state to state (perhaps even county to county within a

state) for services for mentally handicapped students. In

addition, Singer and Irvin (1989) pointed out that at each

stage different types of support for families needs to be

provided. For example, early intervention is needed during

school years, while during the transition to postsecondary

years, supportive employment (e.g., sheltered workshop), and

planning for life after a parent's death are needed.

Child's age. Several studies have revealed that

parents of younger children with handicaps experience less

stress than parents of older children (Beckman, 1983;

Bristol & Schopler, 1984; Donovan, 1988; Friedrich,

Wilturner, & Cohen, 1985; Holroyd & McCarthur, 1986; Minnes,

1984, 1986). For example, Minnes (1984, 1986) found that

parents of preschool children with normal intelligence and

physical handicaps (e.g., with congenital limb deficiency)

experience less stress than parents of same-aged preschool

children with an MH. In contrast, parents of primary

school-aged children with an MH experience more stress than

parents of same-aged, normally intelligent, amputees.

Minnes speculated that these results reflected the greater

dependency and management needs of older children with an

MH. Other researchers maintain that reduced stress during

the early years is related to (a) greater social

acceptability and manageability of younger children, (b)

lack of knowledge and denial regarding the permanence of the






35

disability, (c) greater optimism for the future, and (d) the

availability of more and better services for young children

than for older children and adolescents (Bristol &

Gallagher, 1989; Bristol & Schopler, 1983; Donovan, 1988).

In contrast, Flynt and Wood (1989) found that mothers of

children with a moderate MH grouped along three normative

transition periods (school entry, early adolescence, and

young adult) were not significantly different in perceived

family stress levels.

The results of research concerning the impact of a

child's age on family stress are limited due to

methodological problems, however. For example, some studies

have been conducted using a narrow age range (e.g., Beckman,

1983; Wilton & Renaut, 1986), while others have used a broad

range of ages (Friedrich et al., 1985; Holroyd & McCarthur,

1986; Minnes, 1984, 1986). In addition, the relationship

between parental stress and child's age has been found to

differ according to the severity of the mental handicap

(e.g., Blacher, Nihira, & Meyers, 1987). Blacher et al.

(1987) found that children with a moderate or severe MH

negatively impacted family adjustment early in life, with

this effect remaining stable over time. In contrast, the

impact of children with a mild MH decreased over time.

Finally, in studies conducted by Minnes (1984, 1986),

children with an MH were grouped according to age without

differentiating severity of mental handicap within groups.

In addition to the child's age, life-cycle stages can be

described in terms of the ages of the parents.








Parent's age. A few researchers have addressed the

relationship between parental age and family functioning

variables (Minnes, 1988b). Beckman (1983) found no

significant differences in levels of stress reported by

older and young mothers of children with an MH. In

contrast, Flynt and Wood (1989) found that older mothers of

children with a moderate MH perceived lower stress than did

younger mothers. A possible reason for the discrepancy

between these studies is that Beckman used a restricted age

range (19 to 39 years), whereas Flynt and Wood (1989) used a

broader one (22 to 60 years). Minnes (1988b) suggested that

older, well-established parents with a handicapped child may

respond quite differently than younger, less experienced

parents because they may have more resources (e.g.,

financial) and larger support networks. Nevertheless, Flynt

and Wood (1989) did not find a significant difference in the

coping strategies used by younger and older mothers.

Economic status. Responding to the needs of children

with an MH is costly (Fewell, 1986). Often, a parent may be

unable to seek employment outside the home because of the

management needs of a child with an MH. It is thus not

surprising that levels of stress and adaptation reported by

families of children with an MH have been associated with

the family's socioeconomic status (SES). In spite of this,

Gallagher, Beckman, and Cross (1983) reported lack of

consensus within the literature concerning the impact of SES

on family stress and adaptation. Several researchers have








found that families with limited socioeconomic resources

report higher levels of stress than families with more

resources (e.g., Bradshaw & Lawton, 1978; Friedrich &

Friedrich, 1981; Kazak & Marvin, 1984). Families led by

single parents also report more stress than two-parent

families, with the differences being attributed to limited

financial resources (Beckman, 1983; Salisbury, 1987).

Additional support for the economic impact on families

raising a child with special needs has been found in studies

with middle socioeconomic status (SES) families (e.g., Dyson

& Fewell, 1986; Frey et al., 1989). Surprisingly, Minnes

(1984, 1986) found that higher SES parents reported greater

stress than parents from middle and lower SES.

Level of stress in families of young children with

handicaps was studied by Dyson (1991) after controlling for

SES in a sample of matched family pairs (SES and family

structure--one versus two--parents) of families of children

with and without handicaps. The results revealed that

families of children with an MH experienced elevated levels

of stress that are independent of economic and social

conditions. Similarly, Flynt and Wood (1989) and Dunst,

Trivette, and Cross (1986) found that SES did not affect

family stress in families of children with a moderate MH.

These findings are supported by Byrne and Cunningham (1985),

who found no significant effects for SES on family stress in

their review of the literature.








It is difficult to compare results across studies

since researchers vary in their definition and measurement

of SES. For example, some researchers have used published

measures of SES (e.g., Dyson, 1991; Flynt & Wood, 1989),

while others have created their own scales (e.g., Dunst et

al., 1986; Wikler, 1986a, 1986b). These contradictory

findings may also be further explained by cultural

differences (Minnes, 1988b). For example, the findings may

be a result of differences in the expectations of a

particular culture or ethnic group regardless of their SES.

Ethnicity. Few researchers have considered the

relationship between the parent's and the child's race or

ethnic background on family functioning. The studies that

have been conducted have reported conflicting results

(Minnes, 1988b). For example, Friedrich (1979) used race

(white versus nonwhite) as one of several demographic

variables used to predict adaption in mothers of children

with handicaps (physical, motor and mental). In this study,

race was not found to contribute significantly to the

mothers' feelings of capability in coping with the handicap.

In contrast, Flynt and Wood (1989) found that African-

American mothers of children with a moderate MH reported

lower perceived stress levels than did Caucasian mothers.

Moreover, the African-American mothers reported greater

utilization of coping strategies centered around intrafamily

resources and garnering social support. Recently, Hanline

and Daley (1991) examined the coping strategies and family








strengths of Hispanic, African-American, and Caucasian

families of preschool children with and without varying

disabilities. They found that for each group refraining (an

internal coping strategy) tended to be more predictive of

family strengths (an internal resource) than did the use of

social supports. Other significant within-culture findings

from this study will be discussed in the sections on coping

strategies and resources included in this chapter.

Interactions among Stressor Variables

It is possible that interactions among the parent and

child variables provide better explanations for the family's

adaptation than individual variables (Mines, 1988b).

Possible interactions include that of the child's age and

severity of mental handicap. That is, if variables were

further analyzed it is possible that the effect of severity

of mental handicap is present with younger age children and

not older children. Similarly, the age of the parent may

affect the adaptation to the severity of the handicap. For

example, older parents of children with more severe MH may

experience more stress. Typically, studies in this area

have not addressed these possible interactions (Minnes,

1988b). Therefore, further research in this area is needed.

Summary

In sum, research suggests that family stress and

adaptation are related to differences in the characteristics

of the child and the parent. However, for the studies that

considered child and parent variables, the results were

equivocal. Typically, differences across studies can be

explained by methodological differences (i.e., sampling and






40

instrumentation differences) (Minnes, 1988b). Research also

suggests that demographic variables alone are insufficient

for explaining the range and complexity of familial

adaptational responses. In the next section, the roles of

resources, coping strategies, and perceptions as mediators

of stress and promoters of family adaptation are discussed.

Resources and Adaptation (B Factor)

McCubbin and Thompson (1987) defined family resources

as those qualities the family has available for dealing with

stressors. Family resources include (a) personal resources

of individual family members, (b) internal or systemic

characteristics of the family that contribute to its role

structure and organization, and (c) social support received

from extended family, friends, professionals, and agencies

outside the immediate family. The first two types of

resources are considered internal, whereas the third is

considered external resources. Those aspects of family

functioning that have been cited as resources include mutual

assistance, family and self (individual) esteem, family

integration, sense of mastery, expressiveness, financial

resources, social supports and the degree of independence in

the family (Burr, 1973; McCubbin, 1989; McCubbin, Cauble, &

Patterson, 1982). Family resources can either exist already

or be developed and utilized to help manage stressors and

demands (McCubbin, 1989).

Internal resources. Internal resources are those

intrafamily characteristics that are used to respond to







41

crisis situations, cope with normative and nonnormative life

events, and promote growth and development in all family

members. They are also referred to as family strengths in

the literature (Dunst et al., 1988; Minnes, 1988b; Turnbull,

1990). Several researchers have established the importance

of a family's internal resources in the management of stress

in families with a child with an MH (Abbott & Meredith,

1986; Friedrich et al., 1985; Mink, Nihira, & Meyers, 1983,

1988; Minnes, 1984; Nihira, Meyers, & Mink, 1980). In

samples of families of children with a mild, moderate, or

severe MH, internal resources were also found to be related

to family adaptation for two types of stressors: general

life stressors and those specific to the care of the child

with an MH (Minnes, 1988b). Families of children with more

severe MH reported more difficulties in general family

adaptation, as measured by the Family Environment Scale

(Moos & Moos, 1981), which measures relationships, personal

growth, and maintenance functions (Mink et al., 1983).

Similarly, families of children with mild and moderate MH

demonstrated higher levels of internal resources (e.g.,

cohesiveness) and lower levels of stress associated with

specific child stressors (e.g., dependency and management

needs) than did families with children with severe MH.

However, compared to families of school-aged children with

normal intelligence, families of children with an MH were

not statistically different in the levels of two types of

family internal resources: family pride--which refers to






42

respect, trust and loyalty--and family accord--which refers

to sense of mastery and competence (Abbott & Meredith,

1986). Furthermore, Abbott and Meredith (1986) found that

most of the families of children with an MH in their sample

reported that they developed additional strengths as the

result of the presence of the child with an MH. For

example, families with children with an MH reported growing

stronger, closer, and more unified. Moreover, about 40% of

the parents of children with an MH reported increased

personal resources (e.g., compassion, patience) as a result

of having a child with an MH.

Results from these studies should be viewed cautiously,

however, because self-report questionnaires were used and

the contributions of a variety of factors, including gender

of the child and parental age were not considered (Minnes,

1988b). Furthermore, limited aspects of each factor (e.g.,

internal resources) are typically assessed. For example,

Abbott and Meredith (1986) found that adaptation was

significantly related to family pride and accord, which are

two specific internal resources measured by the Family

Strengths Scale (Olson, Larsen, & McCubbin 1982). It cannot

be assumed that these relationships hold for other types of

internal resources (e.g., good health).

External resources. In addition to resources within

the immediate family systems, the types and availability of

external resources (e.g., family, friends, professionals)

also are thought to contribute to family adaptation (Dunst








et al., 1988; Hanson & Hanline, 1990; Minnes, 1988b). For

example, Wikler (1986b) found a strong positive relationship

between social support and a family's adaptation to stress

resulting from dealing with life transitions and critical

events associated with having a child with an MH. Social

support relationships have also been found to be related to

parental attitudes regarding children with handicaps (Crnic,

Friedrich, & Greenberg, 1983; Crnic, Greenberg, Ragozin,

Robinson, & Basham, 1983; Orr et al. 1991) and reduce levels

of stress (Dyson & Fewell, 1986). In addition, social

supports and the number of sources of support have been

shown to be positively related to family integrity, parental

perceptions of child's functioning, parent-child play

opportunities, and child behaviors and outcomes (Dunst,

Trivette, & Cross, 1986).

While the majority of studies support a positive

relationship between social support and adaptation, the

results from these studies have been challenged. Dunst et

al. (1988) recently stated that interventions directed at

increasing additional external resources (e.g., involving

more professional supports) may be less effective than

interventions geared at strengthening internal family

resources. Disagreement in the literature also exists

regarding the labeling of social supports (as either

external resources or coping strategies). Some believe that

social support should be considered a coping strategy

because it refers to assistance given to or sought by








families (Turnbull & Turnbull, 1990); others view social

supports as one set of resources that can be mobilized

(coping behavior) when internal resources fail, subside, or

are depleted (Schilling, Gilchrist, & Schinke, 1984).

Another important issue is whether social support is a state

or a trait (Farran, Metzger, & Sparling, 1986). If it is a

state, then the family's level of social support can be

determined by counting the number of people and agencies

with which they interact. If social support is a trait,

then the family's satisfaction with level of support is the

critical measure.

These debates highlight the many difficulties in

studying the concept of social support. Clarification of

these issues may have significant implications for practice.

For example, if a family lacks social support (state) then

the most appropriate intervention might be to increase the

number of people with whom families interact. In any case,

it is evident that additional research on social support is

necessary (Gallagher & Bristol, 1989).

Coping strategies. Research suggests that, in addition

to resources, coping strategies can be crucial for families

of children with an MH (Gallagher & Bristol, 1989; Mink,

Nihira, & Meyers, 1983, 1988; Minnes, 1988b). McCubbin and

Patterson (1982) included coping strategies in the B factor.

However, close examination of the definition of coping

strategies by McCubbin and Thompson (1987) shows that it is

difficult to delineate the parameters of this concept. In






45

general, coping strategies are considered behaviors (covert

or overt) that are directed at acquiring and allocating

resources (McCubbin & Thompson, 1987). McCubbin and his

associates have identified five types of coping strategies:

(a) direct action to reduce the number or intensity or both

of demands, (b) direct action to acquire additional

resources not already available to the family, (c)

maintaining existing resources so they can be allocated and

re-allocated to meet changing demands, (d) managing the

tension associated with ongoing strains, and (e) changing

the meaning of situations to making them more manageable.

Thus, as was the case for resources, coping strategies can

be classified into external strategies (acquiring social,

spiritual, and formal supports) and internal strategies

(reframing and passive appraisal). It is important to note

that internal coping is subsumed under perceptions in the

Double ABCX model. Thus, it appears that one type of coping

strategy, external resources, is subsumed under the B

factor, while the second, internal resources, is subsumed

under the C factor.

Perceptions or Family Definitions of the Stressful Event
(C Factor)

Perceptions (C) are internal coping strategies that

involve thinking about a stressful situation, either to

change one's definition of the event (making the situation

feel less stressful or to make it solvable) or to facilitate

acceptance of the event (Turnbull & Turnbull, 1990). The

three major types of internal coping strategies are passive






46

appraisal, refraining, and spiritual support. While passive

appraisal and refraining clearly constitute internal

activities, spiritual support is considered both an internal

coping strategy and external resource (Fewell, 1986). If

spiritual support is conceptualized as use of a religious

framework or beliefs, then it is an internal strategy.

However, if spiritual support is operationalized as external

relationships that come from membership in a religious

organization, then it is an external resource.

It is difficult to clearly ascertain the relationship

between perceptions and adaptation in families of children

with an MH because so few studies have been conducted

exploring this relationship. Perhaps this is due to the

paucity of instruments that have been developed to measure

this factor (Wikler, 1986b). In one study, Orr et al.

(1991) found that families of children with an MH that

parental stress was significantly related to positive

appraisal of the child's condition. Rabking and Steuning

(1976) examined the relationship between the family's

perceptions to characteristics of the child with an MH and

other types life stress (e.g., optimistic outlook). Their

results revealed that general perceptions of life stressors

were positively related to the family's perceptions of the

child's handicap. However, if families were typically

optimistic about life, then they were optimistic about the

child's future.








Resources and Coping Strategies

In addition to the relationships found between

resources and adaptation, and between coping strategies and

adaptation and resources, a significant relationship between

resources and coping strategies has also been established

(Abbott & Meredith, 1986; Hanline & Daley, 1991). For

example, Hanline and Daley (1991) found that two types of

internal resources (family pride and accord) were

significantly positively related to the coping strategies

employed by families of children with disabilities (D) and

without disabilities (ND). Moreover, these relationships

were found to vary across ethnic groups. For both Hispanic

and African-American families, a significant positive

relationship was found between family pride and refraining

for D and ND. Family pride and seeking spiritual support

were significantly related for Hispanic ND families. Family

accord and seeking spiritual were significantly related for

Hispanic D families. For Caucasian D families, minimizing

reactivity was negatively related to internal resources

(family pride and accord). Minimizing reactivity or passive

appraisal was also negatively related to family pride for

Caucasian ND families. It is important to note that these

findings may be influenced by social class, as primarily

middle class families were included. In addition, other

possible contributing child characteristics were not

considered (e.g., severity of MH).








In sum, significant relationships have been found

between resources, coping strategies, perceptions, and

adaptation. A significant relationship between internal

resources and coping strategies has also been established.

Nevertheless, controversy exists regarding the parameters of

resources and coping strategies, as well as the

conceptualization of the stressor and adaptation.

Limitations of the Double ABCX Model

Discrepancies in the literature regarding the

interpretation of the Double ABCX model exist. These

discrepancies stem from disagreement regarding the

parameters of the resource, coping strategies, and

perception factors. McCubbin and Thompson (1987) defined

the bB factor in terms of capabilities. They also made a

distinction between two types of capabilities: resources

and strengths (qualities families have) and coping behaviors

or strategies (how families employ these qualities).

Perceptions were defined as the family's appraisal of the

stressor. McCubbin and Thompson (1987) also suggested that

passive appraisal is one type of coping strategy. Thus,

there seems to be two types of coping strategies, one

subsumed under the bB factor and another under cC. To this

end, some authors have interpreted the Double ABCX model as

having a bB factor, BC factor, and a cC factor (e.g.,

Gallagher & Bristol, 1989).

Opinions regarding these distinctions differ (Dunst et

al., 1988; Gartner, Lipsky, & Turnbull, 1991; Orr, Cameron,








& Day, 1991; Singer & Irvin, 1989). Orr et al. (1991)

supported the conceptualization of the B factor as

comprising both resources and the use of those resources,

and the C factor as reframing (perception). In contrast,

Dunst et al. (1988) state that internal and external

resources should be separate. Still others argue that

McCubbin and his associates have overlooked the fact that

the concepts of internal resources, external resources, and

coping skills overlap considerably and may need to be

recombined in a different manner or be considered one factor

(Crnic, Friedrich, & Greenberg, 1983). Still others assert

that internal resources, external resources, and coping

skills are independent factors, with perceptions being

subsumed under coping skills (Singer & Irvin, 1989).

In addition to the discrepancies regarding the

conceptualization of internal and external resources, coping

skills, and perception, the distinction between the stressor

and adaption factors is unclear. In a review of studies

investigating the Double ABCX model, significant differences

in the definitions of these factors and their assessment was

found. For example, the level of stress has been

alternatively reported as an indicator of the stressor

event/pile-up demands (aA) (Orr et al., 1991), as an

adaptational response (cC) (Wikler, 1986b), and as the end

state, adaptation (xX) (Frey, Greenberg, & Fewell, 1989).

Furthermore, the interaction of ecological contexts with

factors is unclear. While some state that the Double ABCX

does consider the role of the ecological context (Gallagher






50

& Bristol, 1989), others suggest that the ecological context

is not considered (Crnic, Friedrich, & Greenberg, 1983).

Thus, despite the empirical support for the Double ABCX

framework, the limitations noted suggest that modified

versions of this model may be necessary to explain family

functioning. Minnes (1988b) calls for the use and

refinement of the Double ABCX model to facilitate analysis

of the complex multiple factors contributing to family

stress and coping. Indeed, several authors have proposed

modified versions of the Double ABCX model for studying

families of children with an MH. In these modified versions

the factors are redefined or recombined, or both, to

highlight different contributions of the Double ABCX model.

Modified Versions of the Double ABCX Model

Model of Stress. Adaptation, and Ecological Context

Drawing from the literature in the areas of stress,

individual coping, and ecological influences on development

and functioning, Crnic, Friedrich, and Greenberg (1983)

proposed the Model of Stress, Adaptation, and Ecological

Context. To the ABCX model, Crnic et al. added the concept

of ecological context, that is, the conditions and the

setting in which the family exists. These authors

emphasized that the family's coping resources are affected

by the ecological contexts of the individual family members

(e.g., classroom, work), whole family (e.g., home), peer

groups and societal institutions (e.g., public schools,

government agencies), as well as by the interactions between

and within these contexts (as illustrated in Figure 3).

This model departs from the Double ABCX model in that coping








strategies and resources (internal and external) are

conceptualized as one factor, coping resources. However, by

collapsing these two concepts into one, they have failed to

distinguish between family resources or qualities and coping

strategies or family activities (McCubbin & Thompson, 1987).

For example, one may have a large extended family (external

resource), but may not ask them for assistance (coping

strategy). Furthermore, the individual contributions of

resources and coping strategies to family adaptation as well

as the interaction between them become difficult to

delineate in this model.



-STRESS



ECOLOGICAL CONTEXT / COPING RESOURCES


ADAPTATION

positive-----negative


Figure 3. Model of Stress, Adaptation, and Ecological
Context (Crnic, Friedrich, & Greenberg, 1983).


For example, in studies by Frey, Greenberg, and Fewell

(1989) and by Crnic and Greenberg (1990) examining the

functioning of parents of children with various handicapping

conditions (ages 0 to 6), several issues are evident. Both

studies supported the overall predictive power of the

Stress-Adaptation model. However, the predictive power of

the individual components subsumed under the coping








resources factors (i.e., those measuring family social

network, parent belief systems, and coping styles) varied

significantly. Furthermore, significant interactions were

found between individual components, including social

network and belief system. Frey et al. (1989) found that

the most powerful predictor of parental adaptation to stress

was the parental belief system. Moreover, Crnic and

Greenberg (1990) found that mothers' and fathers' appraisals

(i.e., ability to reframe and actively appraise stressors)

were strongly related to family adjustment. In addition,

mothers reporting high levels of passive appraisal also

indicated more stress and poorer family adjustment. The

same was not true for fathers. This supports the necessity

for the distinction within the "coping resources" factor.

Although Greenberg and his colleagues did not address that

issue, they did suggest that their findings were consistent

with regard to the role of coping appraisal in the

individual (person-specific) coping model proposed by

Folkman and Lazarus (1984).

Crnic and Greenberg (1990) also found mothers and

fathers of children with an MH on the Brief Symptom

Inventory (which measured individual stress) were similar to

the normative sample. Higher levels of stress were reported

by parents of male children with an MH than by parents of

female children with an MH. This suggests a difference in

adaptation due to the child's gender. One interesting

finding was that the parents of both female and male








children with communication difficulties reported higher

levels of stress than children with adequate expressive

language skills. Thus, it may not be severity of an MH, but

certain features of an MH (e.g., communication) that affect

family adaptation. It is possible that children with a more

severe MH have more communication difficulties. Crnic and

Greenberg unfortunately did not examine that possibility.

Social network was, however, positively related to family

adjustment, while the child's age was not.

Stress-Adaptation Model

In the Farran, Metzger and Sparling (1986) modified

version of the Double ABCX model, the basic factors were

redefined and relabeled. They proposed the Stress-

Adaptation Model which was originally'used in a study on the

effects of parental unemployment on family functioning.

This model has three factors: (a) Demands, (b) Adaptive

Capacity, and (c) Adaptation (see Figure 4). The first

factor, Demands, is composed of both the event stressorr)

and the changes associated with the event. The event in

this model is comprised of the various features of the

stressor event (e.g., severity, duration), contemporaneous

events or periodic crises, or both. Therefore, additional

life stressors and strains experienced by the family are

included in this factor. The Demands factor is composed of

changes associated with the event, including those that

affect personal evaluations of self and future, family

routines, and role within the family. These changes

presumably occur in families of children with handicaps








because of the demands placed on a family as the result of

the handicap (e.g, mother must stay at home). The second

factor, Adaptive Capacity, refers to the ways people define

a stressor event (considered a personality characteristic)

and social supports. Finally, the last factor, Adaptation,

refers to the functioning of the family/parents (physical

and mental health) and the child (physical and mental

health).


DEMANDS
Event
Changes associated
with the Event


ADAPTIVE CAPACITY



IMPACT ON IMPACT ON
ADULTS CHILDREN


Figure 4. Stress-Adaptation Model (Farran, Metzger, &
Sparling, 1986).


Empirical evidence for this model comes from a large

research project examining the effect of head of household

unemployment on family functioning (Farran & Margolis, 1984)

In general, results indicate that individual family member

functioning (e.g., physical and mental well-being) was

related to the adaptive capacity of the individual (e.g.,

prior unemployment status and amount of social support).

Quantitative studies with families of children with an MH

have not yet been conducted. However, researchers have

presented evidence supporting this model in case studies of






55

families with children with an MH. In fact, one of the case

studies documents the family adaptation of one of the

authors, Joyce Metzger, who experienced the death of a

daughter with spina bifida and life with a son with a

seizure disorder and an MH (Farran, Metzger, & Sparling,

1986).

In this model, specific reference is made to separate

parent and child outcomes. This distinction enables the

examination of individual family member outcomes. However,

specific child outcomes (e.g., health, academic achievement)

have not yet been examined with this model. Nonetheless,

the distinction highlights a significant area in need of

further study. Like the other models presented thus far,

changes in a family (e.g., personal evaluation of self and
p
future) and a family's adaptive capacity (e.g., personal

protective factors) overlap in this model. Another

limitation of this model is its linear format. Farran and

Margolis contend that, despite its linearity, the model is

applicable to chronic conditions because, as families

experience chronic strains, they cycle periodically through

the linear pathway. That is, a new cycle is triggered each

time new events occur that demand changes in the family.

Although Farran and Margolis suggest the possibility of an

interaction between components, the model as presented does

not take into account such interactions. Lastly, this model

has only been empirically evaluated with families with an

unemployed parent. However, case studies of families of

children with an MH do support the applicability of this

model. The authors suggested that they are currently








conducting research on the effects of ultrasound diagnosis

on the health outcomes of the parent and the child after

birth (Farran et al., 1986). However, these studies are as

of yet unpublished, and it is not clear how such studies

would lend empirical support to the applicability of the

model to children with an MH and their families served by

the public schools.

Model of Family Stress

Recently, Singer and Irvin (1989) proposed the Model of

Family Stress, a modified version of the Double ABCX family

model that is parallel to a model of stress and adaptation

in individuals presented by Billings and Moos (1984; see

Figure 5). Like the Farran et al. model (1986), this model

of family adaptation is similar to the model of adaptation

to individual stress proposed by Folkman and Lazarus (1984).

Like the Double ABCX model, it maintains that the effects of

the stressors are determined in part by intrafamily and

community resources and in part by the family's appraisal

and coping response.

However, several departures from the Double ABCX model

are noteworthy. First, resources have been separated into

internal and external resources, with both separated from

coping response. Thus, there are two resource factors or,

two "B" factors, in this model. Although not labeled as

such by the authors, a distinction between intra-family

resources (Bl) and community resources (B2) is useful for

discussing this model. Second, the family stressors are

defined as problems that are directly related to the child's

handicap and can be specific to the child (e.g., behavior






57

problems, ill health, appearance), the family (e.g., social

isolation, money worries), or both. Finally, the model does

not endorse an A to B to C to X path, but, rather, a B to A

to C to X path. That is, intra-family resource and

community resources (Bs) effect the stressors (A). In turn,

the stressor impacts the family appraisal and coping

responses (C). The C factor effects positive or negative

family adaptation (X).


Positive
INTRA-FAMILY RESOURCE ADAPTATION



^ FAMILY APPRAISAL [
STRESSOR and -
COPING RESPONSES


\ / _____ Negative
COMMUNITY RESOURCES ADAPTATION


Figure 5. Model of Family Stress (Singer & Irvin, 1989).


Singer and Irvin (1989) emphasized that, although the model

is described sequentially, each factor could affect every

other factor. For example, intra-family resources (Bl)

might influence not only the stressor (A), but also the

community resources (B2), family appraisal and coping

response (C), and adaptation (X).

To date, studies employing the whole Model of Stress

(i.e., examining all the factors) have not yet been

conducted. A few researchers examined the relationship

between specific factors (e.g., social support-family








adaptation), however. For example, Cooley, Singer, and

Irvin (1989) found a significant relationship between

extensive social support, in the form of volunteer friends

for the children with a severe MH, and positive family

adaptation. Singer and several colleagues (Singer & Irvin,

1989; Tordis & Singer, 1991) have concentrated primarily on

describing the implementation of this framework in applied

settings (e.g., special education preschools). Given

efforts to apply this model, it is likely that the results

of evaluation studies will be available in the future.

Summary

In summary, several models are available for use with

families of children with an MH. The models are all similar

in that they are based on the Double ABCX framework.

However, they differ in the ways in which the variables of

the model are defined. These modified versions also have

limitations. The field of family stress and adaptation is

at a point where more research on the family variables that

are emphasized in existing models, as well as proposals of

new models, is necessary (Wikler, 1986a). Clearer

conceptualizations and additional studies of the

relationship between factors are necessary because of the

need for empirically sound, family-focused interventions.

Family Assessment Instruments

There are several reviews of the instruments that have

been designed to assess whole family functioning (Grotevant

& Carlson, 1990; Halvorsen, 1991; Touliatos, Perlmutter, &








Strauss, 1990). Typically, these reviews are limited to

published instruments; the myriad experimental scales

created to serve a specific purpose are not included. Both

published and experimental instruments often lack documented

evidence of their reliability and validity. Moreover,

reviews consistently highlight the lack of consistency and

the variety among these instruments.

In a recent editorial, deGruy and Dickinson (1991)

proposed that the main problem of family assessment stems

from the complex and abstract nature of the concept of

family, which is neither defined nor measured easily.

Nevertheless, they suggested that clinicians and researchers

accept the multiplicity of definitions and simply be

explicit about whichever definition they use. Another

source of problems in assessing families arises from the

diversity of theoretical orientations to families. Again,

deGruy and Dickinson suggested that clinicians and

researchers accept the multiplicity of theoretical

orientations and discuss their choices.

Once a definition and theoretical orientation is

presented, the method of assessing the family presents

another set of problems. Family assessment can be conducted

in a variety of formal and informal methods, including

observations, interviews, and self-report. Researchers of

families with children with an MH argue passionately about

the method or combination of methods that should be used in

this area (e.g., Turnbull & Turnbull, 1990, Stoneman &








Brody, 1990). However, as deGruy and Dickinson (1991)

stated, "just as it is not necessary to select a single

definition, neither is it neither is it necessary to select

a single method as the best for collecting data" (p. 18).

Therefore, researchers should select the best method for

their particular research question and theoretical

orientation.

Following this advice, instruments used to

operationalize the factors in the present study are the

self-reports scales of one individual (parent) about

selected aspects (stressors, strains, internal resources,

coping strategies, and adaptation) of the family as a whole.

The self-report scales used are theoretically-based (Double

ABCX model).

Self-report questionnaires are often necessary because

they are time- and cost-efficient and provide an insiders

view of the family. Several problems concerning both rater

competence and psychometric properties arise when self-

report questionnaires are used, however. First, self-report

formats are vulnerable to social desirability and other

response biases, and may be limited by the subjects own-

self/other awareness (Grotevant & Carlson, 1990). Second,

self-report scales are limited by their reliability and

validity. Unfortunately, few family assessment instruments

with adequate reliability and validity data are available

(Dunst et al., 1988; Grotevant & Carlson, 1990; Halvorsen,

1991; Minnes, 1988b; Stoneman & Brody, 1990). Nonetheless,






61

effort was taken to use scales with demonstrated reliability

and validity.

The scales used in this study are discussed in the next

chapter. In addition, a description of the sample and data

collection method are presented in Chapter 3.














CHAPTER 3
METHODOLOGY

The research method used in this investigation was

designed to facilitate examination of the relationship

between stressors, internal resources, coping strategies,

and adaptation in families of children with a mental

handicap (MH). The data collection procedures, description

of the sample, instrumentation, and statistical analyses are

presented in this chapter.

Data Collection

Permission to conduct the study, as well as a list of

eligible schools, were obtained from the Supervisor of the

Programs for Mentally Handicapped Students. Teachers of

students with an MH and their school principals were

informed about the study in a letter (see Appendix A).

Packets containing a cover letter (see Appendix B),

questionnaires (printed in a booklet and titled "Family

Scales"; available upon request) and a self-addressed,

stamped return envelope, were delivered by the examiner to

each participating school. A total of 648 packets were sent

to the parents) or guardian(s) by the children's teachers.

The packets were sent via the normal means of teacher-parent

communication (e.g., hand delivered, through bus personnel,








etc.). A total of 139 parents/guardians (21.45% return

rate) returned the questionnaire. Due to the low response

rate and to address concern about possible bias in the

original sample, an additional 32 parents/guardians were

randomly selected by the Supervisor of Mentally Handicapped

Programs of Hillsborough County, Florida, based on scheduled

educational placement staffing meetings. The questionnaire

was administered (100% return rate) by the author at various

schools after the parents/guardians attended the staffing

meeting. The combined sample consisted of 171 families.

Description of the Sample Obtained

Comparison of the Mail-in and Visited Samples

The means for the mail-in sample (return rate = 21.43%)

were compared to the means for the visited sample (return

rate = 100%) to examine the biasing effect of choosing to

respond to the mailed survey. Independent sample t-tests

showed no significant differences on any of the demographic

variables between those parents/guardians that mailed in the

scales (N=139) and those visited (N=32) by the author (see

Table 1). With the exception of the Behavior Control (BC)

scale, the means of the mail-in and the visited samples on

the family scales were also not significantly different (see

Table 2). Therefore, no systematic pattern of response bias

was found. Consequently, the data from the mail-in and

visited samples were pooled for the remaining analyses. The

demographic characteristics of the combined samples and mean

scaled scores are presented in the next section.










Table 1

Scalh Comnarisons for Denmoaranhic and


Stressor Variables


Mail-In Visited Combined
(N=139) (N=32) (N=171)
Variable Mean SD Mean SD t-test Mean SD




CGENDER(F) 0.42 0.49 0.44 0.50 0.21 0.42 0.50

PGENDER(F) 0.89 0.37 0.91 0.29 0.20 0.89 0.37

CAGE 9.26 2.30 8.44 2.20 1.84 9.11 2.30

PAGE 2.85 1.01 2.81 1.03 0.18 2.84 1.01

ETHNIC(C) 0.63 0.49 0.56 0.50 0.66 0.61 0.48

ETHNIC(H) 0.09 0.29 0.13 0.37 0.53 0.10 0.30

TMH 0.29 0.45 0.41 0.50 1.31 0.31 0.46

SPMH 0.13 0.34 0.16 0.37 0.40 0.13 0.34


*P < .05










Table 2

Scale Comparisons for Main-In and Visited Samples


Mail-In Visited Combined
(N=139) (N=32) (N=171)

Variable Mean SD Mean SD t-test Mean SD


AS

RF

SSP

MF

PA

FS

RS

SS

FWB

SFS

SD

FSTRESS

FSTRAINS

PS

CM

RL

AR

AI

BC

GF


23.88

31.19

14.22

13.63

13.59

33.76

37.63

8.90

24.86

3.68

11.55

14.25

9.72

1.92

2.08

2.26

1.88

2.03

1.72

1.84


5.54

5.31

3.97

3.43

2.78

8.07

10.96

2.74

9.44

1.60

3.15

10.65

9.49

0.51

0.49

0.49

0.63

0.54

0.49

0.54


24.00

31.90

15.53

14.12

14.31

34.84

39.34

9.40

24.97

3.60

10.78

13.29

10.69

1.77

1.97

2.13

1.67

1.97

1.54

1.68


5.45

4.10

3.58

2.86

2.62

5.75

10.36

2.11

8.11

1.60

2.96

11.57

9.56

0.51

0.45

0.38

0.54

0.45

0.29

0.53


0.11

0.72

1.70

0.75

1.34

0.72

0.80

0.98

0.05

0.29

1.25

0.45

0.52

1.53

1.18

1.50

1.81

0.65

2.02*

1.52


23.90

31.32

14.47

13.73

13.73

33.96

37.95

9.00

24.89

3.67

11.40

14.07

9.90

1.89

2.06

2.23

1.84

2.02

1.67

1.81


5.51

5.10

3.93

3.34

2.76

7.68

10.84

2.64

9.18

1.60

3.12

10.80

9.49

0.51

0.48

0.48

0.62

0.53

0.46

0.54


*p < .05








Subjects

A description of the entire sample (N=171) is presented

in Table 1. Regarding the race of the families studied, 61%

were Caucasian, 10% were Hispanic, and 29% were African-

American. Of the 171 responding parents, the majority were

mothers (89% female) [PGENDER (F)]. Fathers comprised 11%

of the sample. For convenience, the ages of the parents

(PAGE) were placed in six categories (1=20-24 years, 2=25-34

years, 3=35-44 years, 4=45-54 years, 5=55-64 years, and 6=65

years and over). The mean age group of the parents was

2.84, which indicates the average age of the parents in the

sample was between 25 to 44 years. The gender of the

children [CGENDER (F)] with an MH whose parents participated

in the study was 42% female and 58% male. The average age

of the children (CAGE) was 9.11 years. The children were

enrolled in programs for students with intellectual

functioning within the Educable Mentally Handicapped (EMH)

range (56%), Trainable Mentally Handicapped (TMH) range

(31%), and Severe/Profound Mentally Handicapped (SPMH) range

(13%).

Instrumentation

The four page questionnaire booklet (The Family

Scales), which contained all the measures used (available

upon request), was given to the parents/guardians to

complete. Permission to use each of the published scales

was obtained from the respective authors. The instruments

were all self-report, paper-and-pencil questionnaires. The






67

Family Scales was composed of five scales: (a) Demographic

variables, (b) Family Crisis Oriented Personal Scales

(F-COPES) (McCubbin, Larsen, & Olson, 1982), (c) Family

Inventory of Resources for Management (FIRM) (McCubbin,

Comeau, & Harkin, 1981), (d) Family Stressors (McCubbin &

Patterson, 1982) and Family Strains (McCubbin & Patterson,

1982), and (e) Family Assessment Device (FAD) (Epstein,

Bishop, & Baldwin, 1982). Responses from the five scales

were used as measures of the four factors as defined in this

study: (a) Demographic and Stressor variables, (b) Internal

Resources, (c) Coping Strategies, and (d) Adaptation. The

Demographic Scale was created by the author to gather

demographic information. The other four scales are

published instruments.

Demographic Variables

Demographic variables were obtained from responses on

Scale 1. On Scale 1 of the Family Scales booklet, the

respondent (parent/guardian) was asked to identify the

following: (a) child's chronological age, (b) child's

gender, (c) child's educational program (severity of mental

handicap), (d) child's ethnicity, (e) respondent's age, (f)

respondent's gender, and (g) respondent's ethnicity.

Economic status was operationalized as the score from the

Sources of Financial Support Index (FSI) on the FIRM. This

index reflects the sense of stability in family income

(e.g., dependence on public assistance). The FIRM is

reviewed in the next section.








The severity level of the child's mental handicap

(i.e., educable, trainable, severe/profound) was determined

from the parent's report of the special educational program

in which the child currently receives instruction, as

placement is based on severity level. The child's severity

of mental handicap was used as a measure of the stressor

variable.

In addition to the stress of having a mentally

handicapped family member, it was hypothesized that the

additional life stressors (i.e, pile-up stressors) faced by

the family are also stressor variables. Family stressors

include events experienced by the family in the past year

(e.g., addition of member, changes in the work situation,

illness, deaths) that can render the family susceptible to

the impact of a subsequent stressor or change. These events

were measured by 10 items on the Family Stressors Index

(McCubbin et al., 1982) and were presented as the first 10

items of Scale 4 in the Family Scales booklet.

Family Inventory of Resource Management (FIRM)

The Family Inventory of Resource Management (FIRM)

(McCubbin, Comeau, & Harkin, 1981) is a 69-item, 4 point

Likert-type scale (ranging from Not at All to Very Well)

that assesses the family's social-psychological, community,

and financial resources along personal, family system, and

community dimensions. The instrument contains four scales:

(a) Family Strengths I: Esteem and Communication (FS),

which reflects family esteem, communication, mutual








assistance, optimism, and problem-solving ability; (b)

Family Strengths II: Mastery and Health (RS), which

includes a sense of mastery, mutuality, and physical and

emotional health of family members; (c) Extended Family

Social Support (SS), which taps the help and support given

and received from relatives; and (d) Financial Well-Being

(FWB), which measures the perceived financial efficacy. The

FS, RS, and FWB were used to assess internal resources; the

SS was used to measure coping strategies.

Social desirability and financial status were also

assessed using the FIRM. The Social Desirability Scale (SD)

of the FIRM is based on the Edmond's Scale of Marital

Conventionalization (Edmonds, 1987). The Sources of

Financial Support (FSI), an additional index on the FIRM,

gathers information regarding property ownership and sources

of income (e.g., alimony, welfare). The FSI has moderate

reliability (alpha = .44) and "reflects the sense of

stability and esteem associated with income" (McCubbin &

Comeau, 1987, p. 146).

Reliability. Information regarding the reliability of

the FIRM is reported in the manual (McCubbin & Thompson,

1987). The internal consistency for the total score is .89

(Cronbach's alpha). The internal consistency for the FS,

RS, SS, and FWB scales are .85, .85, .63, and .85,

respectively. Thus, adequate internal consistency for the

overall FIRM and three of the four major scales have been

established. The SS Scale demonstrated a lower internal








consistency. Examination of the inter-scale correlations

revealed low correlations among the scales, with the

exception of a moderate correlation between the SD and

family strengths scales (FS and RS) (mean r = .43). Test-

retest correlations are not yet available for the FIRM.

Validity. According to the manual, the four scales of

the FIRM were identified via a factor analysis of data

obtained from 322 families with a chronically ill child.

The convergent validity of the FIRM was also supported by

the moderate correlations between the FIRM total score and

the Family Environment Scales (FES), which was designed to

assess family characteristics and relationships (Comeau,

1985). Significant positive correlations were observed

between the FIRM scales and the FES family environment

dimensions of cohesion, expressiveness, and organization;

negative correlations were observed between family conflict

and four FIRM scales (FS, RS, SS, and FWB). The construct

validity of the FIRM has also received some support. In a

recent study with a sample of diverse families, the FIRM

Extended Family Social Support (SS) scale was minimally

correlated with the same dimension of the Family Stress and

Social Inventory (FSSI) (Halvorsen, 1991), a new measure of

family stress and social support. Halvorsen (1991)

suggested that low correlation appeared to be the result of

a greater emphasis on the FSSI regarding the availability of

extended family members (e.g., grandparents) than on the









FIRM. Halvorsen did not compare the FSSI with other FIRM

scales.

Several studies have been conducted that establish the

predictive validity of the FIRM with families of children

with chronic illness (Comeau, 1985) and mental handicaps

(Orr, Cameron, & Day, 1991). Comeau (1985) also found that

the total FIRM score was a good predictor of positive health

outcomes in chronically ill children. Similarly, the total

FIRM score was found to be significantly related to the

total stress score on the Parenting Stress Index (Abidin,

1986; Orr et al., 1991).

Family Strains Index

The Family Strains Index (FSTRAINS) (McCubbin,

Patterson, & Wilson, 1982) was designed as part of the

Family Indices of Regenerativity and Adaptation (FIRA)

Series (McCubbin & Thompson, 1987). The 10 items of the

FSTRAINS represent the areas of Intra-Family Strains, Work

Strains, and Chronic Illness Strain on the original Family

Inventory of Life Events (FILE; McCubbin, Patterson, &

Wilson, 1979).

The FSTRAINS is a self-report instrument that assesses

the pile-up of life events experienced by a family (the A

factor of the Double ABCX model). In spite of the fact that

it was developed as a measure of stress, an examination of

the items on the FILE that made up the FSTRAINS seems to

suggest that in actuality it measures strains that deplete

family resources. For example, the index assesses conflict








between husband and wife, conflict among and with children,

financial hardships and the strains of caring for an ill

member.

Limited psychometric information is available regarding

the FSTRAINS. Moreover, studies using the FSTRAINS have not

yet been published. McCubbin and Thompson (1987) reported

the Family STRAINS Index has an internal consistency of .69

and adequate validity, based on a correlation of .87 with

the original FILE. Unfortunately, correlations between the

FSTRAINS and a newer version of the FILE (Form C) are not

yet available. Nonetheless, the correlation with the

original FILE is important, given the FILE's well-

established internal consistency (e.g., r = .81), as well as

adequate predictive and concurrent validity. Thus, the

FSTRAINS has been proposed as a useful research tool;

however, the limited psychometric information warrants a

cautious interpretation of the information from this

measure. To increase confidence in the FSTRAINS results

obtained, an internal consistency estimate for the sample in

this study was generated (see Chapter 4).

Family Stressor Index

Like the FSTRAINS, the Family Stressor Index (FSTRESS)

(McCubbin, Patterson, & Wilson, 1982) was created as part of

the FIRA. It is composed of 10 items from the FILE.

McCubbin et al. have suggested this is an adequate tool for

research, primarily because of the reduced number of items.

The FSTRESS records those life events and changes that can








render a family vulnerable to the impact of a subsequent

stressor or change. The index assesses the addition of a

member, changes in the work situation, illness, and deaths.

McCubbin and Thompson (1987) reported that the FSTRESS index

correlates .60 with the FILE. Because an internal

consistency coefficient was not available in the manual, one

was generated for the sample in this study (see Chapter 4).

Family Crisis-Oriented Personal Evaluation Scale (F-COPES)

McCubbin, Larsen, and Olson (1982) created the Family

Crisis Oriented Personal Evaluation Scales (F-COPES) to

assess the C factor in the Double ABCX model. The F-COPES

is a 30 item, 5-point Likert type scale (1 = strongly

disagree; 5 = strongly agree) that assesses the problem-

solving approaches used by families/in response to a

stressor.

The F-COPES was selected for the assessment of the

coping strategies because of its emphasis on resources and

strengths. In addition, it has been used in studies of

diverse cultures and of families with disabled children

(Hanline & Daley, 1991).

The normative sample for the F-COPES scale consisted of

2,740 husbands, wives, and adolescents. Through factor

analytic procedures five factors were identified in support

of the five individual scales: (a) Passive Appraisal (PA),

which measures the family's ability to accept problematic

issues minimizing reactivity, (b) Reframing (RF), which taps

the family's capability to redefine stressful events to make








them more manageable, (c) Acquiring Social Support (AS),

which refers to the family's ability to actively engage in

acquiring support from relatives, friends, extended family,

and neighbors, (d) Mobilizing Family Resources (MFR), which

measures the family's ability to seek out community

resources and accept help from others, and (e) Seeking

Spiritual Support (SSP), which focuses on the family's

ability to acquire spiritual support. The first two scales

measure covert or internal behaviors and the latter three

scales measure overt or external behaviors. In addition to

scores for the individual subscales, a total score can be

obtained.

Interpretation of the F-COPES scale scores requires

comparison of obtained scores to the normative data listed

in the manual. Means and standard deviations for the

normative sample are provided. However, the F-COPES total

score does not appear to reflect the overall level of family

coping because of the mutual exclusiveness of some coping

strategies. For example, active strategies (e.g., those

measured by the scale Mobilizing Family to Acquire and

Accept Help) are added to passive strategies (e.g., Passive

Appraisal). Thus, the total F-COPES scores may not be

reflective of a strength in one set of strategies as opposed

to another. The F-COPES also has adequate reliability and

validity.

Reliability. The authors reported Cronbach's alpha

coefficients ranging from .64 to .84 for the five scales,








and a total scale alpha of .86. Hanline and Daley (1991)

found minimal relationships among the scales of F-COPES in

their study of culturally diverse families of young children

with and without disabilities. This finding supports the

independence of the scales. Test-retest reliability over a

period of four weeks was reported to be .81 for the combined

scale and .61 to .95 for individual scales. A significant

correlation of .47 between the SS and MF scales with a

sample of families of children with an MH was reported by

Minnes (1988a).

Validity. Several researchers have established the

predictive validity of some of the F-COPES Scales (Barnett,

Hall & Bramlet, 1990; Hanline & Daley, 1991; Minnes, 1988a;

Orr et al., 1991). Typically, these authors focused on

particular F-COPES scales as opposed to all the F-COPES

scales. For example, the Passive Appraisal (PA) scale was

positively related to the stress associated with caregiving

demands and everyday life stressors, as measured by the

Parental Stress Index (Abidin, 1986), in parents of

preschool children (Barnett et al., 1990) and in parents of

children with a moderate mental handicap (Orr et al., 1991).

These findings suggest that for these groups of families,

parents accepting problematic issues and minimized

reactivity reported lower levels of stress.

Minnes (1988a) used the AS, SSP, and MF as measures of

external family resources in a study of families with

children with a severe MH. The results revealed that SSP








was not related to internal family resources, child and

parent characteristics (e.g., age, gender), and parental

stress. A significant negative relationship was obtained

between SSP and parental lack of personal reward (r = -.39).

Researchers using the F-COPES in their family stress

research provide further information regarding the contents

of the F-COPES scales. For example, the relationship

between F-COPES and family internal resources was

investigated in a study of families with three-year-old

children with and without disabilities from African-

American, Hispanic, and Caucasian families (Hanline & Daley,

1991). The results revealed that the RF scale was

negatively related to family pride (r = -.50) in Hispanic

families, but positively related for African-American

families of children with and without disabilities. Family

accord was found to be negatively related to the PA and AS

in Hispanic families of children with disabilities. PA was

negatively related to family accord in Caucasian families of

children with and without disabilities. AS was positively

related to family pride in Caucasian families of children

with disabilities. This study highlights that the

predictive power of the F-COPES may vary across ethnic

groups.

In another study of families of children with an MH,

the Reframing scale of the F-COPES was found to have a

significant, moderate correlation with the total FIRM score

(Orr et al., 1991). This relationship may be the result of








the internal or covert nature of Refraining and family

resources. Significant relationships were not found between

other F-COPES and FIRM scales.

In sum, the F-COPES appears to have adequate reliability

and validity. Of the F-COPES scale, Passive Appraisal (PA)

appears to be significantly related to both family internal

resources and family stress. However, these relationships

varied across three cultural groups. Differences between

families of children with and without disabilities were also

evident.

Family Assessment Device (FAD)

McCubbin and Thompson (1987) listed several scales that

could be used to measure family adaptation. Following their

suggestion, the Family Assessment Device (Epstein, Baldwin,

& Bishop, 1983) was used in this study. The FAD is grounded

in the extensively researched McMaster Model of Family

Functioning (Westley & Epstein, 1969). This model is

congruent with family stress theories in which individual

members are viewed as part of a larger family system. This

model is based on the assumption that the function of the

family unit is to provide a setting for the development and

maintenance of the family member on social, psychological,

and biological levels.

The FAD is a 60 item, 4-point Likert scale which covers

six areas: (a) problem solving (PS), which refers to the

family's ability to solve problems at a level that maintains

effective family functioning, (b) communication (CM), which








taps the extent to which family members express themselves

clearly and directly to the person for whom the message is

intended, (c) roles (RL), which measures whether the family

has established patterns or behavior for handling a set of

family functions and the extent to which tasks are clearly

and equitably distributed, as well as carried out

responsibly by family members, (d) affective responses (AR),

which addresses the extent to which individual family

members are able to experience appropriate affect over a

range of stimuli, (e) affective involvement (AI), which

describes the extent to which family members are interested

in and place value on each other's activities and concerns,

and (f) behavioral control (BC), which measures the way in

which families express and maintain standards for the

behavior of its members. A score for each area can be

calculated, as well as an index of general functioning (GF),

which assesses the overall health/pathology of the family.

Reliability. The FAD has acceptable internal

consistency for each of its scales and the overall score

(GF). In a recent study by Kabacoff, Miller, Bishop,

Epstein, and Keitner (1990), the F-COPES was administered to

a large sample of nonclinic (i.e., without mental or

physical illness) families and families of medical and

psychiatric patients. The mean internal consistency

coefficients for the scales ranged from .65 to .84, with a

median coefficient of .74. Similar results were obtained by

Epstein et al. (1983). Moderate between-scale correlations

(from .40 to .60) were obtained among all the FAD scales








(Epstein et al., 1983). This finding suggests the scales

overlap to some degree. However, partial correlations

between the dimensions approached zero when GF was held

constant. Thus, the variance between the six scales is for

the most part accounted for by the variance that each shares

with GF. Moderate test-retest reliability coefficients for

the seven scales (mean r = .70) are reported in the manual

(Epstein, Bishop, & Miller, 1982). Epstein et al. also

reported that the FAD is free from social desirability bias,

as evidenced by the small correlations with the Marlow

Crowned Social Desirability Scale (correlation range: -.06

to .44; mean correlation = -.19 for 7 scales).

Validity. The FAD has been carefully developed in

terms of item generation and item- and scale-level analysis.

It appears reliable and has some measure of concurrent

validity with at least one other measure of family

functioning, The Family Unit Inventory (FUI). The FUI was

developed using factor analysis and measures family

Integration and Coping. Correlations of above .50 were

found between five of the six FAD scales and the FUI

Integration scale (Miller, Epstein, Bishop, & Keitner,

1985). Low to moderate significant correlations were found

between the FUI Coping scale and each of the FAD scales. In

addition, a substantial relationship was found between the

FUI Integration scale and the FAD General Functioning (GF)

scale. A moderate relationship was found between GF and the

FUI Coping scale.








The discriminant validity of the FAD was also

established by Miller et al. (1985). In a sample of

psychiatric and Lupus patients, the FAD discriminated the

healthy from unhealthy families on six of its seven

dimensions (Epstein et al., 1983). Behavior Control was the

one scale that did not differentiate the families. Clinical

cut-off scores on the FAD have modest degrees of sensitivity

and specificity, and it is easy to administer and score.

However, the FAD did not correlate with FACES II, a well-

established measure of family adaptability and cohesion.

Whether this is a problem with the FAD or the FACES II is

not clear.

The constructs of the FAD need further study, as the

authors do not provide sufficient description with regard to

the similarities and differences with related constructs

from other instruments. The authors acknowledge that the

Behavior Control (BC) scale needs more work. Moreover,

sample sizes for the validation studies are small and

relatively homogeneous.

Data Analysis

In this section statistical analyses used are

described. Because of the confusion in the literature (see

Chapter 2) about what constructs are being measured, a

factor analysis of the 20 scales administered was conducted.

The following is a list of acronyms:

1. AS Acquiring Social Support

2. RF- Refraining

3. SSP Seeking Spiritual Support

4. MF Mobilizing Family to Acquire and Accept Help









5. PA Passive Appraisal

6. FS Family Strengths I: Esteem & Communication

7. RS Family Strengths II: Mastery & Health

8. FWB Financial Well-Being

9. SFS Sources of Financial Support

10. SS Extended Family Social Support

11. SD Social Desirability

12. FSTRESS Family Stressors Index

13. FSTRAINS Family Strains Index

14. PS Problem Solving

15. CM Communication

16. RL Roles

17. AR Affective Responsiveness

18. AI Affective Involvement

19. BC Behavior Control

20. GF General Functioning

Scree plots were examined to determine the number of

factors. A Varimax rotation was used for scale

interpretation. The standardized scales were then summed

and used to create variables for the regression analysis

(i.e., to test the model). To answer the three research

questions, the following three regression analyses were

conducted: (a) internal resources were regressed on

stressors, (b) coping strategies were regressed on

stressors, and (c) family adaptation was regressed on

stressors, internal resources, and coping strategies. The

results from the factor analysis and regression analyses are

presented in the next chapter.














CHAPTER 4
RESULTS

Results are presented in this chapter in two sections.

In the first section the descriptive statistics for each

measure are presented, as well as the reliability data for

the Family Stress and Family Strains Indices. The second

section presents the results of the various correlational

procedures used to answer the main hypotheses of this study.

It should be noted that the results reflect family

functioning as perceived by the parent or guardian who

completed the scales.

Description of Obtained Scale Scores

The descriptive statistics for all variables are

reported in Table 2. Also reported in this table are

results of t-tests of the means for the Mail-In and the

Visited samples across each of the scales. The intent of

the t-tests was to examine possible bias in the Mail-In

sample due to the low return rate. The Visited sample had a

100% return rate. No significant differences were found on

19 of the 20 measures, indicating little or no bias in the

Mail-In sample. The means for the Family Assessment Device

(FAD) scales (AS, RF, SSP, MF, and PA) are consistent with

those of other studies of nonclinical families (Epstein,

Baldwin, & Bishop, 1983). These results are also consistent








with those for families of children with chronic illnesses

for the Family Inventory of Resource Management (FIRM), and

for families of adolescents with and without psychiatric

disorders for the Family Crisis-Oriented Personal Evaluation

Scale (F-COPES), the Family Stress Index (FSTRESS), and the

Family Strains Index (FSTRAINS; McCubbin & Thompson, 1987).

Because of the potential impact of socioeconomic status

on the results of the study, an analysis of each of the

items of the Sources of Financial Support (SFS) Index was

conducted. The percentage of respondents responding No and

Yes on the SFS items are presented in Table 3.


" Table 3

Percentages for Responses on the Sources of Financial
Support Index


Percent Percent
Item No Yes

We have money coming in from our
investments. 81 19

We depend entirely upon financial support
from welfare. 21 79

We depend almost entirely upon income from
alimony and/or child support. 13 87

We depend almost entirely on Social
Security income. 87 13

One or more working parent is currently
unemployed. 26 74

We own land or property besides our place
of residence. 87 13

We own (are buying) a home (single family,
condominium, townhouse, etc.). 48 52








The majority of the respondents indicated that their

families did not have money coming in from investments

(81%). A large percentage indicated that their families

depended upon financial support from welfare (79%) and upon

income from alimony and/or child support (87%).

Approximately half of the families owned or were buying a

home (52%); however, most of the families did not own

property besides their place of residence (87%). In

addition, 74% of the families had one or more working parent

who was currently unemployed.

Internal consistency. Due to the limited psychometric

information available for some of the scales used in this

study, the internal consistency for the FSTRESS and the

FSTRAINS was computed using Cronbach's alpha. Internal

consistencies for the other scales were reported in Chapter

3. A coefficient of .54 was obtained for the FSTRESS and

.74 for the FSTRAINS, indicating that these scales have

modest to good internal consistency.

Factor analysis. A principal components analysis was

conducted to determine the correlational structure of the

variables investigated. Principal component analysis is a

factor extraction method commonly used in the family

psychology literature (e.g., McCubbin & Thompson, 1987). In

addition, Gorsuch (1983) points out that component analysis

and common factor analysis differ substantially only when

the number of variables is small and the communalities are

low. The intercorrelations of all the 20 scales are








reported in Appendix C. Examination of the scree plot led

to the retention of three factors. The first 10 eigenvalues

were: 6.06, 2.55, 2.00, 1.24, 1.17, .99, .89, .87, .70, and

.61. Varimax rotation was used to facilitate interpretation

of these three factors. The unrotated factor loadings are

reported in Appendix D.

The loading of each scale on the rotated factors are

shown in Table 4. While any cutoff for interpreting factors

is arbitrary, Gorsuch (1983) suggested that a loading of .3

should be used only with a minimum sample size of 175. On

the other hand, an absolute loading of .4 can be interpreted

with an analysis based on 100 individuals. Because of the

sample size in this study, the more conservative cutoff (.4)

was used as the salient loading. All the scales of the FAD

(Problem Solving [PS], Communication [CM], Roles [RL],

Affective Responsiveness [AR], Affective Involvement [AI],

Behavior Control [BC], and General Functioning [GF]) load

negatively on the first factor. In contrast, the Passive

Appraisal (PA) scale of the F-COPES and the Family Strengths

I: Esteem & Communication (FS) scale of the FIRM load

positively on this factor. The first factor reflects Family

Adaptation, as each of these aspects of family functioning

contribute to the maintenance of the family system. Four of

the FIRM scales (Family Strengths I: Esteem & Communication

[FS], Family Strengths II: Mastery & Health [RS], Financial

Well-Being [FWB], and Social Desirability [SD]) load










Table 4

Loadings of Each Scale on the (Varimax Rotated) Principal
Components


Factors
Variables Family Internal Coping
Adaptation Resources Strategies Communality


AS

RF

SSP

MF

PA

FS

RS

SS

FWB

SFS

SD

FSTRESS

FSTRAINS

PS

CM

RL

AR

AI

BC

GF


-0.24

0.00

-0.07

0.04

0.45*


0.40*

0.21

0.21

0.16

0.16

0.02

0.02

-0.06

-0.62*

-0.75*

-0.65*

-0.88*

-0.81*

-0.80*

-0.87*


-0.10

0.37

0.25

-0.10

0.06

0.62*

0.71*

0.32


0.79*

0.23

0.67*

0.16

-0.62*

-0.33

-0.32

-0.51*

-0.12

0.00

0.06

-0.29


0.78

0.48*

0.58*

0.76*

-0.19

0.37

-0.04

0.54*

-0.00

0.09

-0.19

-0.39

-0.14

0.01

-0.17

0.02

-0.16

0.11

-0.09

-0.10


0.68

0.38

0.40

0.59

0.24

0.68

0.55

0.44

0.65

0.08

0.48

0.19

0.40

0.49

0.70

0.68

0.82

0.66

0.64

0.85


*Loadings above .40.






87

positively on the second factor, while the FSTRAINS and the

Roles (RL) scale of the F-COPES load negatively. It is

interesting to note that the RL and FS scales load on both

the first and second factors. The second factor seems to

reflect Internal Resources, because these measures assess

internal family capabilities (e.g., self-esteem, sense of

mastery, sense of financial well-being).

With the exception of the Passive Appraisal (PA) scale,

all of the scales of the F-COPES (Acquiring Social Support

[AS], Refraining [RF], Seeking Spiritual Support [SSP], and

Mobilizing Family to Acquire and Accept Help [MF]) load

positively on the third factor. Interestingly, the Extended

Family Social Support (SS) scale of the FIRM, which

McCubbin, Comeau, and Harkins (1982) state measures internal

resources, loads positively on the third factor. The third

factor was termed Coping Strategies because these scales

assess overt and covert behaviors that occur in response to

stressors. In sum, results of the factor analysis revealed

three factors reflecting: Family Adaptation, Internal

Resources, and Coping Strategies.

Factor Intercorrelations

Two main purposes of factor analysis are (a) construct

validation and (b) scaling. Of interest in this study is

the former, which is consistent with typical family

assessment research. In school psychology and family

assessment research, factors are routinely rotated

(obliquely or orthogonally) to facilitate meaningful








interpretation of the underlying correlational structure

among psychological tests, but not for the scaling of

factors. After using the factor analysis for trait

validation, factor scores can be created by summing

variables to obtain potentially correlated factors, even

when the rotation method was orthogonal. For example, Crnic

and Greenberg (1990) subjected the Parenting Daily Hassles

scale to factor analysis with orthogonal rotation. Three

factors with eigenvalues exceeding 1.00 were obtained.

These factors, when summing the variables with high

loadings, were moderately intercorrelated, with correlations

from .41 to .62. The factor scores were then used in

hierarchical regression analyses to examine the extent to

which life stressors and daily hassles predicted children's

behavioral status, parenting status, and family functioning.

Varimax rotation in factor analytic studies of two of the

instruments used in this study (the FIRM and F-COPES) have

also been used for validation purposes, but not for scaling

(McCubbin & Thompson, 1987).

For the regression analyses conducted in this study,

scores on the variables for the three factors were obtained

by summing (subtracting when negative on the factor loading)

the standardized values of the scales that had absolute

factor loadings exceeding .40. Using a factor loading of

.40 as the cutoff criteria is typical of family assessment

research (e.g., Crnic & Greenberg, 1990; McCubbin &

Thompson, 1987). The intercorrelations of the three factors

are presented in Table 5. All factor intercorrelations were








significant, indicating that they share a fair amount of

variance.


Table 5

Correlations of Internal Resources. Coping Strategies, and
Family Adaptation

Family Internal Coping

Adaptation Resources Strategies

Family Adaptation 1.00
Internal Resources 0.62** 1.00
Coping Strategies 0.32* 0.41* 1.00

* R < .01

**R < .001

Note: From the Varimax rotation, factors were created by
summing the standardized variables with loadings exceeding
.40.


Stressors and Internal Resources


A simultaneous regression analysis was conducted using

Internal Resources as the dependent variable. Independent

variables were general life stressors, as measured by the

FSTRESS, and various demographic variables, such as child's

age (CAGE), parent's age (PAGE), child's gender (CGENDER),

parent's gender (PGENDER), and severity of mental handicap

(viz., Severe/Profound Mental Handicap [SPMH], Trainable

Mental Handicap [TMH], and Educable Mental Handicap [EMH]).

An initial regression analysis was conducted with the

interaction of CAGE and severity of MH and the interaction

of CGENDER and severity of MH, as suggested by Minnes

(1988b). The interaction terms were not significant (F (4,

152) = .60, p > .05). The interactions of severity of MH






90

with CAGE (df = 2) and severity of MH with CGENDER (df = 2)

were also nonsignificant. Therefore, the regression

analysis was conducted without the interactions. The

reduced model for the main effects is presented in Table 6.

Results of this analysis reveal a significant

(negative) relationship between CAGE and Internal Resources.

In contrast, PAGE and Internal Resources were found to be

positively related. Another significant relationship was

observed between the family's sources of financial support

(SFS) and Internal Resources. Significant relationships

were not found between Internal Resources and the following

variables: CGENDER, ethnicity (Caucasian, Hispanic, and

African-American), severity of mental handicap, PAGE, and

Family Stress Index (FSTRESS).

Stressors and Coping Strategies

A second simultaneous regression analysis was conducted

using Coping Strategies as the dependent variable.

Stressors variables (CAGE, PAGE, CGENDER, PGENDER, Severity

of MH, FSTRESS, and SFS) were the independent variable, as

well as the interactions of CAGE with severity of MH, and

CGENDER with severity of MH. The interaction terms were not

significant (F (4, 156) = .50, p > .05). The interactions

of severity of MH with CAGE (df = 2) and severity of MH with

CGENDER (df = 2) were also nonsignificant. The reduced

model with the main effects is reported in Table 7. A

significant negative relationship was found between FSTRESS

and Coping Strategies. In addition, the conditional mean of

families of children with SPMH was significantly lower than