Citation
The differential impact of parental death on adolescent stress as determined by individual and family coping resources

Material Information

Title:
The differential impact of parental death on adolescent stress as determined by individual and family coping resources
Creator:
Young, Lucia Patat
Publication Date:
Language:
English
Physical Description:
x, 137 leaves : ; 29 cm.

Subjects

Subjects / Keywords:
Adolescents ( jstor )
Child psychology ( jstor )
Coping strategies ( jstor )
Death ( jstor )
Domestic relations ( jstor )
Grief ( jstor )
Life events ( jstor )
Parents ( jstor )
Psychological stress ( jstor )
Questionnaires ( jstor )
Counselor Education thesis, Ph. D
Dissertations, Academic -- Counselor Education -- UF
Greater Orlando ( local )
Genre:
bibliography ( marcgt )
non-fiction ( marcgt )

Notes

Thesis:
Thesis (Ph. D.)--University of Florida, 1996.
Bibliography:
Includes bibliographical references (leaves 124-136).
General Note:
Typescript.
General Note:
Vita.
Statement of Responsibility:
by Lucia Patat Young.

Record Information

Source Institution:
University of Florida
Holding Location:
University of Florida
Rights Management:
The University of Florida George A. Smathers Libraries respect the intellectual property rights of others and do not claim any copyright interest in this item. This item may be protected by copyright but is made available here under a claim of fair use (17 U.S.C. §107) for non-profit research and educational purposes. Users of this work have responsibility for determining copyright status prior to reusing, publishing or reproducing this item for purposes other than what is allowed by fair use or other copyright exemptions. Any reuse of this item in excess of fair use or other copyright exemptions requires permission of the copyright holder. The Smathers Libraries would like to learn more about this item and invite individuals or organizations to contact the RDS coordinator (ufdissertations@uflib.ufl.edu) with any additional information they can provide.
Resource Identifier:
023838144 ( ALEPH )
35779680 ( OCLC )

Downloads

This item has the following downloads:


Full Text












THE DIFFERENTIAL IMPACT OF PARENTAL DEATH
ON ADOLESCENT STRESS AS DETERMINED BY INDIVIDUAL AND FAMILY COPING RESOURCES
















By

LUCIA PATAT YOUNG


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

1996


UNIVERSITY OF FLORIDA LIBRARIES

































Dedicated to my husband, Dr. Michael Young,
my son, David Young, and my daughter, Allison Young
for the support and encouragement
they gave me throughout this endeavor.
I dedicate this also to the memory of
my dear father, Leon Patat,
and to my mother, Amelia Patat.















ACKNOWLEDGMENTS

I wish to thank the following people for their help in completing this dissertation. I am especially grateful to Dr. David Miller for his encouragement, his kindness, and his depth of knowledge.

I wish to thank the other members of my supervisory committee--Dr. James Pitts for his sense of humor and helpful suggestions, Dr. Robert Ziller for the fun, theoretical discussions, and especially my chairperson, Dr. Ellen Amatea, for her leadership.

I am grateful to all of the participants in my study who "talked with me" about their thoughts and feelings regarding the death of their parents, and I am hopeful that their recommendations to others might be shared through the publication of this study.

I am deeply grateful to my family for their

understanding, support, encouragement, and love throughout this entire process.


iii















TABLE OF CONTENTS


page


LIST OF TABLES .........

ABSTRACT ............


. . . . . . . . . . . vii


CHAPTERS


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

Scope of the Problem .... ...........
Theoretical Framework ..........
Need for the Study ..... .............
Purpose of the Study ... .............
Research Questions ..... .............
Significance of the Study ........
Definitions ...............
Organization of the Study ........

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

Introduction ...... ...............
Impact of Parental Death on the Adolescent


Normal Stress Response to Parental Death Pathological Stress Response to Parental


� . . . 1


...... 23

...... 23
....... 23


� � . 26


Death ..........
Theoretical Models of Childhood Bereavement Stress Theories ..............
Life Events Stress Model .......
Transactional Stress Model ...... Adolescent Coping Models .... ...........
Person-Based Coping Resources .......
Development ...... ...............
Appraisal ...... ................
Style ....... ..................
Family-Based Coping Resources
Parent-Adolescent Communication . ...
Family Adaptability ... ...........
Family Cohesion ..... .............
Conclusion ....... ..................


3 METHODOLOGY .............

Statement of Purpose ......... Research Design ...........


.... . . . 57

.... . . . 57
.... . . . 57


o

o









Delineation of Relevant Variables ......
Dependent Variable ...........
Independent Variables .... ...........
Description of the Population ........
Sampling Procedures .............
Sample ........ .....................
Data Collection ...............
Instrumentation . . . . _ . . .
Measures of Demographic Information . . .
Impact of Event Scale (IES) .. ........
Family Adaptability and Cohesion Scales
(FACES-II) ........ ........
Parent-Adolescent Communication Inventory
(PACI) .......... ........
Adolescent Coping Orientation for Problem
Experiences . ........
Adolescent Questionnaire (Appendix F) . .
Hypotheses ........ ..................
Data Analysis ................

4 DATA ANALYSIS AND RESULTS ..........

Hypothesis one ....... ................
Null Hypothesis Two .............
Null Hypothesis Three ............
Null Hypothesis Four ...... .............
Adolescent Use of Coping Strategies ..... Adolescent Questionnaire .... ............
Summary ...................

5 DISCUSSION ........ ...................

Description of Sample ...........
Question One: Relationship Between Stress and Coping . . . . . . . . . . . . . . . . .
Question Two: The Relationship Between Stress and Parent-Adolescent Communication . . . Question Three: Relationship Between Stress and Cohesion ... .......
Question Four: The Relationship Between Stress Adaptability . ..... . .. ..
Adolescent Questionnaire: Coping Behaviors and Advice ....
Limitations of the Study ..... ............
Implications: Counseling, Future Research, and Theory . . . . . . . . . . . . . . . . . .
Summary . . . . . . . . . . . . . . . . . . . .


* . . 58
* . . 58
* . . 58
* . . 60
* . . 61
* . . 63
* . . 65
* . . 66
* . . 66
* . . 66

* . . 69

. . . 71


81

82
84 88 89 90 98 100

101


and


APPENDICES


A LETTER TO SCHOOL COUNSELORS ...........

B PARENTAL CONSENT LETTER ........ . .. .


101

102 104 105 107 108 109 i1
114


116

117


Q

Q
O









C LETTER TO ADOLESCENT ..... ............... 119

D CHILD ASSENT SCRIPT ...... ............... 120

E DEMOGRAPHIC QUESTIONNAIRE .... ............ 121

F ADOLESCENT QUESTIONNAIRE .... ............. 122

G LETTER TO PARTICIPANTS ..... .............. 123

REFERENCES ......... ...................... 124

BIOGRAPHICAL SKETCH ....... .................. 137















LIST OF TABLES


3-1 Comparative Breakdown by Ethnicity for
Population Subsamples ....... .............. 61

3-2 Frequency Distribution of Participant Age . . . . 64

3-3 Breakdown by County for Research Sample . ..... ..64

3-4 Frequency Distributions of Parental Death by
Sex .......... ....................... 65

3-5 Frequency Distributions of Parental Death by
Type of Death ....... .................. ...65

4-1 Means, Standard Deviations, and Ranges of the
Dependent Variable, Stress, and Independent
Variables: Coping, Communication, Cohesion, and
Adaptability ...... ................... ...83

4-2 Hierarchial Regression Analysis of the
Relationships Among Age, Length of Time Since Death, Communication, Cohesion, Adaptability
with Total Stress Score as the Outcome
Variable ........ . .................... ..85

4-3 Hierarchial Regression Analysis of the
Relationships Among Age, Length of Time Since Death, Communication, Cohesion, Adaptability with Intrusion Subscale Score as a Measure of
Stress ......... ...................... 86

4-4 Hierarchial Regression Analysis of the
Relationships Among Age, Length of Time Since Death, Communication, Cohesion, Adaptability with Avoidance Subscale Score as a Measure of
Stress ......... ...................... 87

4-5 Means, Standard Deviations, and Ranges of the
Twelve Patterns of Coping Behavior Were
Computed to Assess Coping Styles Used by
Subjects in this Study ...................... .91


vii








4-6 Beta Weights for Relationships Between
Independent Variables and Coping by Engaging in
Demanding Activity ...... ................ 93

4-7 Beta Weights for Relationships Between
Independent Variables and Coping by Seeking
Professional Support ..... ............... .93

4-8 Beta Weights for Relationships Between
Independent Variables and Coping by Avoiding
Problems ........ ..................... ..95

4-9 Beta Weights for Relationships Between
Independent Variables and Coping by Solving
Family Problems ...... ................. 95

4-10 Beta Weights for Relationships Between
Independent Variables and Coping by Developing
Social Support ....... .................. 97

4-11 Beta Weights for Relationships Between
Independent Variables and Coping by Seeking
Diversions ........ .................... 98


viii















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

THE DIFFERENTIAL IMPACT OF PARENTAL DEATH
ON ADOLESCENT STRESS AS DETERMINED BY
INDIVIDUAL AND FAMILY COPING RESOURCES By

Lucia Patat Young

August 1996

Chairman: Dr. Ellen S. Amatea Major Department: Counselor Education

The death of a parent is considered to be the most

stressful life event for the adolescent. Adjustment to the new demands and life changes resulting from parental loss are accomplished through the use of both personal and family coping resources. Significant relationships were hypothesized between the degree of stress and the use of certain coping strategies, the levels of family cohesion and adaptability, and the style of parent-adolescent communication utilized with the surviving parent. In this study, individual coping theory, family stress theory, and transactional stress theory provide a foundation for assessing adolescent stress.

A sample of 70 parentally bereaved adolescents were

assessed in terms of the above mentioned variables through a series of multiple regression analyses. Outcome assessments








were represented by a total stress score and subscale scores of intrusion and avoidance. No significant associations were found between the variables and the stress scores. However, the degree of stress reported for the sample was found to be moderate to high. Measures of family cohesion and adaptability indicated that the families of these adolescents were perceived to be of a "mid-range" type. Parent-adolescent communication was reported to be open and positive.

A post hoc analysis of the relationships between coping strategies and family cohesion, parent-adolescent communication, length of time since death, gender of adolescent, and gender of deceased revealed some interesting patterns of coping styles. The most frequently reported coping pattern of behavior in dealing with the death was avoidancee" although the majority of coping styles used were direct or problem-solving methods.

In addition to standardized assessments, adolescents

were asked to discuss coping strategies that were helpful to them in dealing with the death and to discuss the advice they would give to a friend who lost a parent. Analyses indicated that the majority of participants utilized the help of social and family support and advised friends specifically "to talk" about their experiences with a friend or family member.















CHAPTER 1
INTRODUCTION


Adolescents report the death of a parent to be the

most stressful event in their lives (Coddington, 1972; Crook & Eliot, 1980; Dise-Lewis, 1988; McNeil, 1989; Raphael, 1983; Siegel, Mesagno, & Christ, 1990). The loss is a harsh separation that is beyond their control, and they struggle not only to reorient themselves to their environment but also to redefine their relationships with their deceased parents. Moreover, the loss of a parent may generate other stressors for the adolescent such as increased family responsibilities, problematic communication with surviving parents, physical or mental illness in surviving parents, social isolation from friends, economic insecurity, and premature separation from parents (Vida & Grizenko, 1989).

The challenge of adapting to these painful

circumstances can be an overwhelming one. Most adolescents report initial grief reactions such as shock, numbness, and disbelief (Raphael, 1983; Silverman & Worden, 1992). They may demonstrate feelings of sadness, guilt, anger, depression, disorientation, and insecurity (Unpublished manuscript, Gapes, C., 1982) and exhibit behaviors of crying, insomnia, learning difficulties, withdrawal, appetite problems, and temper tantrums (Gray, 1987; Harris,









1991; Silverman & Worden, 1992; Vida & Grizenko, 1989). Some adolescents, however, demonstrate severe symptoms over a long period of time. For example, unresolved, suppressed, and chronic grief response may be acted out in behaviors of school delinquency, increased alcohol and drug intake, phobias, suicide ideation, depression, antisocial behaviors, academic failure, stealing, fighting, reckless driving, unwanted pregnancy, and hostility toward the surviving parent (Freudenberger & Gallagher, 1995; Raphael, 1982, 1983). Compulsive caregiving of the surviving parent is another symptom often demonstrated in which the adolescent suppresses personal grief in an attempt to protect the surviving parent from pain. Many times the adolescent is pressured to take on the role of the deceased parent to regulate emotional distress in the family (Bowlby, 1980b; Cheifetz, Stavrakakis, & Lester, 1989).

Some adolescents develop symptoms of psychopathology during the first year of parental bereavement (Kaffman & Elizur, 1973; Kliman, 1979; Raphael, 1982) while others demonstrate a "sleeper effect" reporting symptoms several years after the loss (Harris, 1991; Vida & Grizenko, 1989). A common reaction to parental death is to regress and demonstrate behaviors characteristic of a younger aged child. To guard against regression, some adolescents demonstrate an inability to cry or express emotion for fear of becoming overwhelmed with grief and feeling like a helpless child (Adams-Greenly & Moynihan, 1983). Many







3

younger adolescents appear preoccupied with the death. In contrast, older teens may use denial and avoidance in dealing with the emotional distress (Harris, 1991).

What distinguishes adolescents who do adapt over time to this painful loss from those who do not? Although researchers have shown increased interest in studying adolescent response to parental death, few efforts have been made to explore the differences between adolescents who have greater difficulty adapting to changes created by parental death than those who are able to adapt more readily.

One promising line of inquiry involves examining the personal and family resources adolescents bring to bear in coping with their loss. Jackson and Bosma (1990) propose that there is a strong relationship between coping and selfconcept in adolescents and suggest that personal coping resources such as internal locus of control, high selfesteem, and increased age and cognitive ability help protect the adolescent in stressful situations. They suggest that if the adolescent is able to establish a positive identity, a self-concept as adults will be forged that is capable of coping successfully with stresses arising in adulthood (Jackson & Bosma, 1990). Silverman, Nickman, and Worden's (1992) "child bereavement study" suggests that after parental death, there is a process of adaptation where the self-concept is reshaped to accommodate a new relationship with the deceased by incorporating the loss. A new reality is constructed that maintains a connection with the parent.







4

This cognitive-emotional approach is described as an ongoing process throughout development that facilitates personal coping and accommodation to life changes.

Researchers have found numerous aspects of family life which influence the coping behaviors of bereaved adolescents. These include family communication, cohesion, and adaptability (Adams-Greenly & Moynihan, 1983; Buehler, 1990; Cragg & Berman, 1990; Crosby & Jose, 1983; Gray, 1989; McGoldrick, 1991; Moody & Moody, 1991; Noller, Seth-Smith, Bouma, & Schweitzer, 1992; Patterson, 1988; Patterson & McCubbin, 1991; Saler & Skolnick, 1992; Silverman & Silverman, 1979; Stern, Van Slyck, & Newland, 1992; Siegel et al., 1990; Walker, 1985). The presence of a supportive family environment, including parental warmth, cohesiveness, and organization, has been found to provide positive models for identification (Compas, 1987). Open communication with surviving parents about a parent's death facilitates the ability to mourn (Siegel & Gorey, 1994). In contrast, failure to discuss the death fosters avoidance and denial of the finality of the loss (Becker, 1973).

Although many adolescents appear to have the

resources needed to cope with the loss of a parent and have no need for interventions from mental health professionals, others do not adapt well to parental loss. This study examines the degree of stress due to parental death reported by adolescents and considers whether certain personal and







5

family coping resources may moderate the degree of stress resulting from this loss.

Scope of the Problem

Numerous theories have been proposed explaining the process of adolescent bereavement and adjustment to loss. Models of adolescent bereavement have been developed based on psychodynamic theories of attachment (Bowlby, 1980a; Freud, 1957a, 1957b; Furman, 1974; Parkes & Weiss, 1983), on stage-specific and task-specific processes of grief (Baker, Sedney, & Gross, 1992), on constructivist theory (Klass, 1988; Rubin, 1985; Rizzuto, 1979; Rosenblatt & Elde, 1990; Silverman, & Worden, 1992), and on stress theory (Buehler, 1990; Compas, 1987; Compas, Malcarne, & Fondacaro, 1988; Dise-Lewis, 1988; Horowitz et al., 1984; Horowitz, Wilner, & Alvarez, 1979; Lazarus & Folkman, 1984; Patterson, 1988; Patterson & McCubbin, 1987; Spirito, Stark, & Williams, 1988).

Cognitive developmental theorists relate the

adolescent's ability to view death as irreversible, natural, inevitable, and personal to their cognitive maturity (Baker et al., 1992; Gordon, 1986; Piaget, 1959; Wass & Stillion, 1988). Whereas pre-adolescents may view death as irreversible, the cause of death is frequently misconstrued and death is seen as an event of the very distant future. In contrast, an adolescent's thoughts about personal death can become a threat to the emerging identity of the adolescent. According to Erikson's (1959) theory, the







6

developmental task of adolescence is to separate emotionally from parents and family while establishing an independent identity. During this period of development, the death of a parent may threaten or arrest identity formation, especially in the younger adolescent years where separation has not been completed (Coddington, 1972; Dise-Lewis, 1988; Erikson, 1959; Raphael, 1983; Wass & Stillion, 1988).

One model of adolescent bereavement relates to

specific stages in which individual tasks of grief are to be accomplished over time (Bowlby, 1980; Furman, 1974; Parkes & Weiss, 1983). The assumption is that if specific tasks are not performed, the person is "stuck" and unable to let go of the deceased and move on with their lives. This linear model is grounded in detachment theory where children are encouraged to cut off ties to the deceased parent, withdraw energy toward the lost love object, and reinvest this energy in new relationships (Freud, 1957b; Furman, 1974). Children are seen as having psychiatric symptoms if they dream about the deceased parent or dwell on their relationship with the deceased.

Another model of bereavement, introduced by Baker and colleagues (1992), lists tasks necessary for successful bereavement and places emphasis on cognitive and contextual issues in the grief process. This model provides a shift from the linear stage specific model and suggests that bereavement might be completed over time. This model incorporates the need for family support and identity






7

development and is an alternative to pathology-based models that identify individuals as being "stuck" in grieving. It promotes a more adaptive view of grief-related behavior and provides direction for short-term therapy goals. Tasks include understanding the fact that someone has died, accepting and bearing the psychological pain of loss, and reorganizing one's identity with a sense of connection to the deceased (Baker et al., 1992).

Departure by several researchers from this task

oriented model to a cognitive-emotional process model places emphasis on cognitive restructuring of the relationship with the deceased parent (Klass, 1988; Rizzuto, 1979; Rosenblatt & Elde, 1990; Rubin, 1985; Silverman, Nickman, & Worden, 1992). This approach, grounded in constructivist theory, theorizes that this occurs when the child renegotiates or constructs a new relationship with the deceased parent over time rather than detaching or letting go. As the child moves through developmental stages, the relationship with the deceased parent changes. Thus, bereavement is considered an ongoing process of adjustment (Silverman & Worden, 1992).

These theoretical explanations in and of themselves

do not explain the differential adjustment of adolescents to the loss experience. A model that emphasizes the role of cognitive mediational processes is the transactional model of stress (Lazarus & Folkman, 1984) with emphasis on cognitive appraisal and coping behaviors (Buehler, 1990;







8

Compas, 1987; Compas et al., 1988; Dise-Lewis, 1988; Patterson, 1988; Patterson & McCubbin, 1987; Spirito, Stark, & Williams, 1988). This approach examines the total relationship the adolescent has with the environment to include the environmental context in which the death occurs, the style of coping with stress, and individual and family resources available to the adolescent (Compas, 1987). Adjusting to the death of a parent is a continuous reappraisal process where the adolescent acquires new coping behaviors and resources to meet the demands or changes created by the loss (Dise-Lewis, 1988; Patterson, 1988).

The relationship between personal resources and coping behaviors is of interest to researchers in determining the type of coping strategies adolescents use to manage stress. The adolescent's repertoire of personal resources such as inherent traits, abilities, and means utilized to meet demands are knowledge, skills, personality traits, emotional and physical health, self-esteem, and self-concept (Patterson & McCubbin, 1987; Jackson & Bosma, 1990). Three types of coping behaviors have been cited in the literature. First, direct actions to reduce or eliminate demands, referred to as problem-focused coping, are one type. A second type consists of behaviors directed at redefining demands to make them more manageable, termed appraisal-focused coping behaviors. A third type, emotionfocused coping, refers to behaviors directed at managing the







9

tension that is felt from experiencing demands (Patterson & McCubbin, 1987).

Researchers have reported that adolescents who use

more approach-oriented, problem-focused strategies to cope with interpersonal stressors have fewer emotional and behavioral problems, whereas those who use more avoidanceoriented, emotion-focused strategies have more problems and are more likely to adjust poorly when confronted by subsequent life crises and stressors (Compas et al., 1988; Ebata & Moos, 1991; Harris, 1991). Different phases of adolescent development are thought to influence the selection and use of particular coping strategies. Younger teens seem to have fewer resources in coping with parental death. They may report overpowering memories of the deceased and preoccupation with the loss. Frequently, older teens use denial and avoidance to escape from dealing with the death. Stress related symptomatology is prominent with both groups, and grief resolution is compromised with these coping methods (Harris, 1991).

Investigators report that adolescents identify peers as being the most helpful to them after parental death and are more likely to talk with siblings and friends about their feelings than talk with parents (Cragg & Berman, 1990; Gray, 1989; Silverman & Worden, 1992). Others have found that the need for adolescents to fit into a peer group may inhibit expression of grief for fear of being singled out as different from peers (Elkind, 1968; Harris, 1991).







10

The most frequently cited family resources in the

research literature reported to facilitate stress management following parental death are cohesion, parent-adolescent communication, and adaptability (Barnes & Olson, 1985; Buehler, 1990; Cragg & Berman, 1990; Gray, 1987; Olson, Portner, & Bell, 1982; Silverman & Worden, 1992). High family cohesion and social support are reported by adolescents to be protective against depression (Balk, 1991; Gray, 1987; Walker, 1985). There is, however, a systematic tendency for adolescents to describe their families as demonstrating less cohesion and less adaptability than their parents report on these values (Noller et al., 1992).

Clinical literature on childhood bereavement stresses that open communication between parents and children greatly facilitates the mourning process (Bowlby, 1963; Cain & Fast, 1966; Furman, R.A., 1964; Furman, E., 1974, 1983; Kliman, 1979; Miller, 1971; Nagera, 1970) and protects youth from the onset of psychopathology in later life (Cheifetz et al., 1989). The ability of each family member to express thoughts and feelings about the death and not be inhibited by one another's emotions is thought to relieve stress and is considered to be "open communication" (Brown, 1988). Adult studies of childhood parental death report that open communication with the surviving parent around issues related to the death of a parent is an important factor in long-term adjustment and protection against depression. In adolescent studies, researchers find that communication is







11

often disrupted in bereaved families (Silverman & Silverman, 1979). The surviving parent and children often avoid discussing the death to spare each other pain. Parents are generally overwhelmed by their own grief and are unavailable to help their own children grieve effectively. Adolescents are not able to accept parental support because they fear it will compromise their needs for autonomy (Adams-Greenly & Moynihan, 1983; Gray, 1987; Harris, 1991; Raphael, 1983).

Family systems theorists report that silence about the death encourages a closed family system and family communication patterns become distorted and rigid (McGoldrick & Walsh, 1991). The death of a parent creates extreme stress within the family. Anxiety is high and attempts are made to protect oneself and other members from the emotional pain of loss. Instead of communicating feelings to other members of the family for emotional support, members fear upsetting others. There is a clear message that members of the family are not allowed to talk about the death or express feelings (Kerr & Bowen, 1988). Crying is discouraged and emphasis is on "getting on with life" and detaching from the deceased. This closed system is unable to integrate the loss and incorporate grieving into its identity while grief is suppressed and denied. Reorganization for the family becomes rigid and dysfunctional and roles are rigidly maintained (Crosby & Jose, 1983; Davies, Spinetta, Martinson, McClowry, & Kulenkamp, 1986). Members are chosen to fill the role of







12

the deceased, and often a child is chosen to take on the role of the deceased adult.

There is little research in the area of adolescent

adjustment to parental loss and how the impact of the family unit affects reorganization and adaptability after the death of a parent. Gehring and Feldman (1988) define adaptability as the ability for the family system to change its power structure of roles and rules in response to situational and developmental stress. Adjustment is found to be greatly determined by the ability of the surviving parent to maintain consistency and stability in the adolescent's environment (Siegel et al., 1990). The functioning family is defined as a flexible unit of individual members who reorganize after the death where no member is expected to fill the role of the deceased parent. Less functional families are described as being resistant to reorganization and they frequently place a child in the role of the deceased parent (Crosby & Jose, 1983; Davies et al., 1986).

Theoretical Framework

At the heart of an exploration into an adolescent's response to parental loss is the conceptualization of the stress experience. Life event theorists report that it is the quality of the life event and circumstances surrounding it that determine the degree of stress incurred by the event (Pearlin & Schooler, 1978). Horowitz et al.(1993) describe the death of a loved one as a life event that evokes stress response syndromes. They describe stress response syndromes






13

as recurrent intrusive images, ideas, and feelings related to a previous traumatic experience. Oscillations between episodes of intrusion and denial are seen as evidence of incomplete personal schema for understanding death and reacting to this event. Horowitz' explanation of intrusion is based on Freud's (1957b) ego-psychological explanation of compulsive repetition of warded-off ideas and associated emotions as an attempt of the ego to master and assimilate traumatic experiences. The repetitive motive is described by Horowitz as a "completion tendency" or a cognitive motive used to reduce the discrepancy between new information elicited by the life event and pre-existing inner models of meaning or schemata. This discrepancy is resolved through the process of assimilation and accommodation defined by Piaget (1959). As long as the memories associated with the life event are held in the active memory processes, it is likely that the information can be processed and assimilated into pre-existing schemata for completion.

However, traumatic memories might activate dreaded

states of mind that are too emotional, too painful, too out of control, and too hopeless. The memories might then become avoided. Horowitz et al. (1979) developed an instrument, the Impact of Event Scale, which measures intrusion and avoidance in response to a life event. Intrusion is represented by the involuntary awareness of ideas, memories, and emotions associated with the event, and avoidance is represented by the conscious attempt to divert







14

attention away from cognitions and feelings related to the event. The instrument is designed to measure subjective experience in response to a life event over time and has been used in recent research to measure the impact of the death of a parent on adolescents (Harris, 1991).

An application of a transactional model of stress to bereaved adolescents is a dynamic conceptualization that emphasizes the relationship between the adolescent and the environment. According to this theory, stress occurs when the individual perceives demands of a situation as taxing or exceeding available coping resources to meet those demands, especially when the system's well-being is perceived as being at stake. It is neither a characteristic of the environment alone nor of the person alone. Rather it is a relational, process-oriented task between the individual and the environment for cognitive-affective appraisal and coping activities. Cognitive appraisal means that the individual has the knowledge or beliefs about how things work and the ability to evaluate the personal significance of encounters with the environment. Appraisal components serve different functions and are influenced by individual development and sociocultural variables. Primary appraisal evaluates stressful environmental conditions to determine the nature of harm/loss, threat, or challenge. Secondary appraisal evaluates which coping options are available and can be used effectively. Person-environment interactions are constantly






15

evaluated and adjustments are made by coping through this reappraisal process.

Coping is viewed as a problem-focused or emotionfocused planned action that regulates emotional distress and shapes subsequent emotion. Problem-focused coping is an attempt to act on the stressor and change the actual personenvironment relationship by directly acting on the environment or on the self. Emotion-focused coping alters the appraisal component of the mind through avoiding the stressor or by cognitively reframing the meaning of the relationship through denial or distancing, thus making the distressing emotion moot. If the meaning of an event is changed, the emotion will also be changed (Lazarus, 1991b). Younger adolescents are reported to use more emotion-based types of coping such as avoidance, denial, and self-blame, whereas older adolescents are reported to utilize direct action as coping methods (Stern et al., 1992).

In addition to problem and emotion-focused coping skills, the presence of a supportive family environment provides resources for successful adjustment to loss in adolescents (Compas, 1987). The integration of individual coping theory with family stress theory provides constructs to examine individual and family resources for adjustment to stress. The most frequently assessed dimensions of family functioning in family research include family cohesion, adaptability, organization, parent-adolescent communication, and conflict resolution skills (Adams-Greenly & Moynihan,







16

1983; Buehler, 1990; Cragg & Berman, 1990; Gray, 1989; McGoldrick, 1991; Moody & Moody, 1991; Noller et al., 1992; Siegel et al., 1990; Silverman & Silverman, 1979; Stern et al., 1992).

The degree to which family functioning and personal

coping skills are instrumental in determining adjustment to the stress of parental death is unknown. The assessment of personal and family resources has the potential of being a helpful diagnostic tool in assessing those adolescents who may be particularly vulnerable to adverse effects of parental loss.

Need for the Study

Little is known about the resources and coping skills that adolescents use to manage the stress created through parental death (Parkes & Weiss, 1983). Research on adolescent bereavement has typically been presented using standardized measures for depression and psychopathology. These measures fail to depict grief-related symptoms as a normal process of bereavement. Normal bereavement needs to be characterized, including risk factors, and special attention given to developmental stages of adolescent development in relation to grief reactions (Harris, 1991; Fleming & Adolph, 1986)

Bereaved adolescents may be at risk because they do not have independent access to treatment as do surviving spouses. Surviving parents may be overwhelmed by their own







17

grief and may not recognize the need for intervention with their children (Harris, 1991).

Many of the studies have relied on parent reports of

adolescent response to death which precludes measurement of those experiences unknown by the parents (Dise-Lewis, 1988; Elizur & Kaffman, 1982; Silverman & Worden, 1992; Vida & Grizenko, 1989), and a number of studies have shown that parental measures often correlate to a lesser extent with adolescent self-report measures (Swearingen & Cohen, 1985).

Adolescent coping strategies have been measured with predetermined lists rather than open-ended measures and response to stress due to parental death has not been charted over a long period of time (Compas, 1987). Age appropriate coping measurements have not been used to reflect changes in cognitive development and response capabilities with parental death. Thus, there is a need to examine a wide range of adolescents differing in their patterns of adaptation to parental loss.

Understanding the impact of coping resources of

bereaved adolescents could provide important information about the intervention needs of adolescents.

Purpose of the Study

The purpose of this study is twofold. First, the

degree of stress reported by bereaved adolescents will be examined. Second, using a transactional stress model, the mediating influences of the adolescent's personal coping strategies and family life on their stress response will be







18

explored. Three aspects of adolescent family life will be assessed: (a) family adaptability, (b) family cohesion, and

(c) parent-adolescent communication.

Research Ouestions

In this study, the relationships between stress, the quality of family life, and personal coping strategies will be described as they pertain to adolescents who have experienced the death of a parent. Specifically, adolescents will be described in terms of the degree of stress experienced, the types of coping strategies utilized, the level of family cohesion, the level of family adaptability, and the degree of positive parent-adolescent communication. The specific research questions are as follows.

1. After controlling for age and length of time

since the death, how do adolescents' use of certain coping strategies relate to their reported degree of stress?

2. After controlling for age and length of time

since the death, how do adolescents' perceptions of their extent of positive parent-adolescent communication relate to their reported degree of stress?

3. After controlling for age and length of time since the death, how do adolescents' perceptions of the level of family cohesion relate to their reported degree of stress?

4. After controlling for age and length of time

since the death, how do adolescents' perceptions of their







19

degree of family adaptability relate to their reported degree of stress?

Significance of the Study

Because certain person-based and family-based

strategies appear to be useful to adolescents coping with bereavement, focusing on the role they play in mediating adolescent stress would seem like a relevant path to take. Specific types of family coping styles would be relevant to look at as well as personal coping abilities.

Definitions

Family adaptability is defined as the extent to which the family system is able to change its power structure, role relationships, and relationship rules in response to situational and developmental stress (Olson et al., 1992).

Adolescence in this study is defined as the period of development between 12 and 18 years of age.

Adolescent adjustment is determined by the level of stress evident as measured by the Impact of Event Scale after the death of a parent.

Avoidance is a psychological reaction to stress and is part of the stress response characterized by ideational constriction, denial of the meanings and consequences of the event, blunted sensation, behavioral inhibition or counterphobic activity, and awareness of emotional numbness (Horowitz et al., 1979).

Cognitive-affective appraisal is an appraisal based on the knowledge or belief about how things work and the







20

ability to evaluate the personal significance of encounters with the environment (Lazarus, 1991b).

Family cohesion is defined as the emotional bonding that family members have toward one another. Within the Circumplex Model, cohesion is measured by emotional bonding, boundaries, coalitions, time, space, friends, decisionmaking, interests and recreation. (Olson et al., 1992).

Family communication is defined as the extent to

which the adolescent expresses feelings directly and shares factual and emotional information with their parent as measured by the Adolescent-Parent Communication Scale (Barnes & Olson, 1985).

Coping is defined as any activity that regulates

emotional distress and shapes subsequent emotion. It is a planned action that is either problem-focused or emotionfocused (Lazarus, 1991a).

Emotion-focused coping is defined as any activity designed to alter the appraisal component of the mind through avoidance or by changing the meaning of the relationship through denial or distancing (Lazarus, 1991a).

Intrusion is a psychological reaction to stress and is part of the stress response characterized by unbidden thoughts and images, troubled dreams, strong pangs or waves of feelings, and repetitive behavior (Horowitz et al., 1979).

Parent, for the purposes of this study, is defined as any person who is the primary caretaker and legal guardian







21

of the adolescent. This person is not necessarily the biological parent.

Problem-focused copinQ is an attempt to change the

actual person-environment relationship by directly acting on the environment or on the self (Lazarus, 1991a).

Psycholoaical stress is a result of a relationship

where environmental demands tax or exceed resources of the individual and negative emotions are present based on harm/loss, threat, and challenge (Lazarus, 1991a).

Stress response is the psychological response to

stressful life events by intrusion and avoidance (Horowitz et al., 1979).

Transactional theory of stress defines stress as a

particular type of relationship between the person and the environment, in which the individual perceives demands of a situation as taxing or exceeding available coping resources to meet those demands, especially when the system's wellbeing is perceived as being at stake.

OrQanization of the Study

This proposal consists of three chapters. Chapter 2 consists of a review of the related literature. Chapter 3 presents the methodology, which includes a description of the research design and variables, the population, sampling procedures, instrumentation, data collection and research questions. Chapter 4 presents the results of the statistical analyses of the data. Chapter 5 includes a discussion of the results, implications for therapy,







22

limitations of the study, and suggestions for further research.















CHAPTER 2
LITERATURE REVIEW


Introduction

In this chapter, a review and synthesis of the literature relevant to adolescent stress response to parental death will be presented including theoretical perspectives on stress response and personal and family coping resources.

Impact of Parental Death on the Adolescent

The majority of literature available explaining

adolescent stress response to parental death utilizes a developmental lens. Differences in adolescent stress responses are posited to be shaped by the child's emotional and cognitive processing capacities. Piaget (1959), for example, suggested that adolescents reach a stage of cognitive development or formal operational thought when they are able, for the first time, to grasp abstract principles and ideas. This cognitive capacity is thought to introduce a new concept of death where death is understood as an irreversible, natural, inevitable, and personal phenomenon (Baker et al., 1992; Gordon, 1986; Wass & Stillion, 1988). The adolescent becomes capable of making a subjective appraisal of the meaning of death, the demands created by the loss, and the personal resources available to









cope with the loss. According to stress theory, death becomes a stressor when the perception of the demands created by the death are outweighed by the perception of capabilities available to meet the demands.

The impact of parental death is viewed by the

majority of researchers as the most profound event shaping the subsequent development of the adolescent. Erikson (1959) proposed that the primary task of adolescent development is to develop an identity and role in society separate from parents and family. Raphael (1983) suggested that the loss of a father or mother during this stage is considered disruptive to normal developmental tasks, instills a fear of lack of control over life events, and creates a threat to the adolescent's mortality. New demands are placed on the adolescent while adjusting to the loss, and the ability to manage the stress created by the death is contingent upon available personal and family resources. Numerous studies have investigated the effects of parental death on various indices of change in the lives of adolescents.

Family relationships change with parental death as well as perceptions of personal and family identities. Increased family responsibilities, social isolation from friends, lack of communication with surviving parents, economic insecurity, and identity crisis due to premature separation from parents all contribute to the stress created through parental loss.







25

Cragg and Berman (1990) conducted a study of ten

adolescents, six months to two years after parental death. They found role reversals in the families as adolescents assumed more household responsibilities and assumed the role of protecting parents who were preoccupied with grief. Some were pressured by family members to take on the role of the deceased parent to help the family avoid the pain of mourning. Adolescents reported social isolation from friends and felt that only those who had lost a parent could understand their feelings. There was reluctance to talk with the surviving parent about their feelings, and a decrease in socioeconomic status was found with one family.

Social isolation from friends has also been described by Gapes (Unpublished manuscript, 1982) in a study of bereaved adolescents who felt that their emotions were unacceptable to members of their social environment and especially to their peers. The adolescents reported that they did not know what was normal in a grief reaction and were hesitant to show their natural responses for fear of being noticed and considered different or abnormal. The typical response of adolescents was one of "escape" by suppressing emotions compared to their surviving mothers who were described as "preoccupied" with mourning. Adolescents felt deficient in expressing emotions, uncertain about the meaning of death, and unsupported by others.

Gray (1989) studied 50 parentally bereaved

adolescents between 6 months and 5 years after the death and







26

found a lack of communication between adolescents and surviving parents. Adolescents reported better understanding by peers than family members and felt more comfortable expressing their feelings outside the family. Peers were rated as the "most helpful" by 80% of the group. Most reported that they relied on only a single friend or a small group of friends for help. Adolescents valued being treated in usual ways and not as a "different" person because they had lost a parent. Some parents were overwhelmed by their own grief and could not help their children.

Baker et al., (1992) suggested that during the early stage of grief, there is a desire to contain the pain of mourning and limit change as much as possible. They imply that there is an intense self-protectiveness of the bereaved individual or family against the full emotional significance of the loss. It is suggested that those who have the least amount of change in their lives adapt better to the stress of parental loss.

Normal Stress Response to Parental Death

Stress response to the death of a parent has been

described by researchers in terms of behavioral, emotional, and physiological reactions. The initial response to parental death has been described as one of shock, numbness, and disbelief (Raphael, 1982; Silverman & Worden, 1992). Gapes (Unpublished manuscript, 1982) defined the adolescent's typical bereavement response as one of "escape"









and identified responses of guilt, anger, depression, disorientation, and insecurity. Research studies of normal, non-clinical populations have been identified to represent the normal stress response to parental death.

Cragg and Berman (1990) identified several indicators of stress, with a non-clinical study of adolescents whose parents died within the previous 6 months to 2 years, such as decreased family communication, increased household responsibilities, increased nightmares, feelings of being different from their friends, thinking that friends could not understand their feelings, and reluctance to talk to parents about the death.

Silverman and Worden (1992) observed normative

behaviors 4 months after parental death in a sample of nonclinical adolescents as follows: crying, insomnia, learning difficulties, health problems with somatization of feelings, headaches, difficulty concentrating in school, external locus of control, and lower levels of self-esteem.

Gray (1987) reported higher depression scores with

bereaved adolescents compared to nonbereaved adolescents and decreased academic performance for younger adolescents indicating that cognitive functions may regress temporarily during bereavement.

In a review of childhood bereavement literature, Vida and Grizenko (1989) found no specific syndrome associated with bereavement, but found an increase in features such as sadness, crying, irritability, minor depressive syndrome,







28

temper tantrums, withdrawal, insomnia, appetite problems, enuresis, and learning and academic problems. The authors suggested that bereavement studies have failed to establish the boundaries between childhood bereavement, major depression, and adjustment disorder. PatholoQical Stress Response to Parental Death

Raphael (1983) suggested that adolescent grief may become pathological due to ego vulnerability. Fears of regression or loss of emotional control may prevent expression of normal feelings of loss and create ambivalent relationships with family and friends. Consequently, adolescents are thought to act out solutions to inner stress often in the manner in which grief reactions are discharged while grief remains unresolved. In case studies of bereaved adolescents, Raphael (1983) found that patterns of suppressed or inhibited grief response, distorted grief, and chronic grief create behavior problems and maladjustment. These patterns were found to create some of the following outcome behaviors: school delinquency, increased alcohol and drug intake, phobias, suicide ideation, depression, antisocial behaviors, academic failure, stealing, fighting, reckless driving, unwanted pregnancy, other risk-taking behaviors, and hostility toward the surviving parent.

Chiefetz et al., (1989) conducted a clinical study of adolescents, from 11 to 17 years of age, whose parent died 2 years prior to the study. Subjects were referred to an outpatient psychiatric service with symptoms of mild







29

depressive affect with dysthymic disorder. Their depressed mood consisted of loss of interest or pleasure in almost all activities for a period of one year. Elizur and Kaffman (1983) also found evidence of dysthymic disorder in a study of parentally bereaved adolescents of 12 years and older.

In the absence of overt symptomatology, another type of pathological behavior has been observed in older adolescents of the same sex as the surviving parent in the form of supportive and extremely protective behaviors of the surviving parent (Elizur & Kaffman, 1983). The behavior was observed more often during conjoint treatment sessions with the adolescent and surviving parent. Bowlby (1980) described similar behaviors as "compulsive care giving" of the surviving parent where the adolescent exhibits exemplary behavior as a defensive facade in the absence of personal grieving. This type of behavior is associated with prolonged mourning.

Several factors that may contribute to pathological stress response to parental death that are not controlled for in studies of bereavement are pre-existing emotional environmental influences in the family, the circumstances of the death, reactions of the surviving parent, and previous life experiences of the adolescent (Cheifetz et al., 1989). Factors such as parental depression and alcoholism are also likely to be factors associated with a more difficult bereavement.







30

Shapiro (1994) suggests that violent deaths or

suicide of parents as well as previous traumatic experiences may contribute to pathological stress response in adolescents. Trauma reactions may accompany, interfere with, or mask grief reactions. Eth and Pynoos (1985) suggested that in the case of traumatic grief reactions, the process of dealing with the traumatic experience takes priority over dealing with the loss. In a study of a violent death where the child has witnessed the physical mutilation of a parent, the fear aroused by the violent image may interfere with exploration of reminiscences that over the normal course of grief, allow identification with the dead parent (Eth & Pynoos, 1985). Lukas and Seiden (1987) found that when the murderer is a stranger, personal safety and fear interfere with the process of grief. When the murderer is a family member, conflicts of loyalty as well as rage occur. With suicide, the child suffers intense guilt in addition to resentment if the child felt responsible for keeping the parent alive (Lukas & Seiden, 1987).

From a review of the literature, there appear to be distinct features of parentally bereaved adolescents. Studies suggest that there is a threat to adolescent identity as motherless or fatherless adolescents adjust to who they are without one of their parents. The possibility of role reversals exist where adolescents are pressured to take on responsibilities of the deceased in order to









maintain some degree of normalcy in the family. Surviving parents may be preoccupied with grief and unable to take care of their families. Communication problems are reported, and there are conflicting reports in the literature about adolescents' willingness to share feelings with family members or peers about their response to the death. There is evidence that those adolescents who are able to process their grief by talking about feelings and the events surrounding the death are able to grieve successfully. Those who are unable to process their grief and feel socially isolated from family and peers seem to have difficulty coping with the loss.

Stress response in adolescents has been distinguished in the literature through clinical and nonclinical populations. It appears that there is a typical grief reaction to parental death and that it is expressed by initial shock, numbness, and disbelief. Nonclinical population studies are thought to represent normal responses to grief and include behavioral, emotional, and physiological reactions to parental death. In contrast, stress response is usually considered pathological in cases of complicated grief where expression of grief is suppressed, inhibited, distorted, or when grief is expressed in chronic behavior patterns that create problems for the adolescent.

In summary, symptoms of bereaved adolescents are specified as normal or pathological depending on the









intensity of stress, the duration of maladjustment, and the extent of adaptability to change. There is no clearly defined bereavement syndrome of adolescents described in the literature that allows the researcher or clinician to clearly define the boundaries between bereavement stress response, depression, and adjustment disorder.

Theoretical Models of Childhood Bereavement

There are several models of childhood bereavement in the literature that describe cognitions and behaviors relative to normal and pathological reactions to the death of a parent. The models are based on detachment theory, stage-specific and task-specific processes of grief, and constuctivist theory.

Theorists traditionally encouraged the bereaved to

disengage from the deceased and let go of the past (Dietrich & Shabad, 1989; Furman, 1974, 1983; Volkan, 1981; Miller, 1971). This approach was grounded in Freud's (1957b) detachment theory which suggests that in order to successfully resolve grief, a person must gradually withdraw mental energy extended toward the lost love object and reinvest the energy in new relationships. Failure to accomplish this detachment was thought to result in symptoms of psychological problems. This theory was conceptualized from clinical case studies with troubled subjects who were more likely to attach inappropriate emphasis on relationships with the deceased (Silverman et al., 1992). Bereavement was seen as a psychological process that was to







33

be completed or resolved in order to adjust. Other models evolved to describe the process further.

Stage specific models were developed where the grief process was explained by a description of emotions from numbness, to yearning, to despair, and reorganization (Bowlby, 1980a; Parkes & Weiss, 1983). Problems were thought to occur when individuals became "stuck" in a stage and were unable to progress to the next stage until they had accomplished tasks of the immediate stage. Bereavement was considered an orderly, linear process that could be successfully completed in sequential stages. There are limitations in applications to treatment with this model because it merely gives a picture of the process at different points in time and does not specify what is required for the individual to move through the stages.

Task specific models have been cited in the

literature and used by Furman (1974) and Shuchter (1986). A time-specific task model was presented by Baker et al. (1992), which outlined specific psychological tasks to be performed over time where internal attachment to the deceased was considered a healthy recovery, not a sign of pathology. This three-phased model first requires that the child have the cognitive ability to understand death and to feel safe in a secure environment. The middle phase requires the child to accept the pain of loss and reevaluate the relationship to the lost object. During the last phase, a reorganization of the child's sense of identity and of







34

significant relationships is required, which allows for a new sense of identity while acknowledging the loss and maintaining some sense of identification with the deceased. This model gives emphasis to developmental stages of the child and to the importance of the role of the family in helping the child process the grief of parental loss. The drawback in this approach is the tendency to propose overly ambitious treatment goals for bereavement; and although the concept of regression during developmental transitions is part of the theory, the implication is that the process may be completed over time. Also, case studies presented with clinical subjects were of troubled children.

Silverman et al. (1992) conducted a child bereavement study of non-clinical children aged 6 to 17 years who had experienced a parental death. Bereavement was considered a cognitive-emotional process whereby the child continued to remain in a relationship with the deceased. This concept was a shift from the former view of bereavement as a psychological state that ends or from which one recovers. Emphasis was on the negotiation and renegotiation of the meaning of loss over time, rather than letting go of the lost love object. This observation is supported in the literature by Rizzuto (1979) who observed that the process of constructing inner representations of the deceased changes with development and maturation. Rosenblatt and Elde (1990) also found that grief work included maintaining connections with memories of the deceased while integrating these







35

memories into the present and into relationships with others. Family members were found to be helpful to each other through shared representations of the deceased. These studies are grounded in constructivist theory where individuals are regarded as constructing their own reality (Gergen, 1985). The bereaved child is thought to construct an inner representation of the deceased parent through memories, feelings, and actions such that the child remains in a relationship with the deceased. This relationship changes as the child matures and accommodates to the loss, developing a new perspective of the meaning of the relationship with the deceased.

Swress Theories

Stress theorists initially referred to stress as an analogy of Hook's 17th century principles of engineering. Stress was considered the ratio of the force of the external stimulant that put strain on an object to the deformity of the object. Emphasis was on the external environment rather than what was happening in the mind. Cannon (1939) and Selye (1956) viewed stress as the physiological disturbance in response to the strain. Selye later coined the term "general adaptation syndrome" which stated that organismic stress was the same regardless of the type of stressor. Focus was on an extreme response to stimuli with the exact nature of the stressor seen as irrelevant to the response.

Life events theorists looked at specific life

experiences, and inventories were developed to measure "life









change units" (Coddington, 1972) to provide an index of the amount of change resulting from the degree of stressfulness associated with experiencing specific events (Johnson, 1986). Not only were specific experiences held accountable for the degree of stress but the frequency and similarity of stressful events were found to increase vulnerability to having subsequent stress reactions in response to stressors of the same type (Horowitz et al., 1979). A more global theory of stress was developed by Lazarus (1966), which suggested that the stressfulness of events not only depends on occurrence of events but on the child's appraisal of the meaning of the events. This cognitive appraisal approach proposed that specific kinds of information determined if a particular stimulus was considered a stressor. Lazarus & Folkman (1984) described appraisal and coping as mediators of stress emphasizing individual differences in reaction to common stressors. This theory further developed into a cognitive-motivational-relational theory of emotion known as a transactional model of stress which provided a broader conceptualization of the interpretation of life events as potential stressors.

Life Events Stress Model

One of the initial inventories which measured life events as units of change or stress for children and adolescents was Coddington's (1972) Life Events Record. The amount of social readjustment required by specific events determined the degree of stress incurred. In a study of









3,526 children ranging from preschoolers to senior high school students, Coddington found a strong relationship between the amount of life change experienced and the age of the child, with older children and adolescents experiencing higher levels of life change than younger children. The death of a parent was more than 20 units higher than the next highest stressor for the participants.

Dise-Lewis (1988) developed a comprehensive measure

of life stress for children based on life events and coping strategies generated by child subjects. The amount of life changes experienced by 681 normal children, the perceived stressfulness of these changes, and the types of coping strategies used by the children to manage stress were incorporated into the Life Events and Coping Inventory. Parental death was rated as the most stressful event in the lives of the subjects. The researcher found that children begin to set a pattern of stress responses by early adolescence and that stress scores correlate well with established measures of anxiety, depression, psychosomatic symptoms, and behavior problems.

Horowitz et al. (1979) studied subjective stress in response to death and constructed a scale, the Impact of Events Scale, to measure stress response as it is expressed through intrusion and avoidance. Intrusion includes involuntary recurrent intrusive images, ideas, and feelings related to the death experience. Avoidance was described as the voluntary, conscious attempt to divert attention from







38

the images, ideas, and feelings related to the experience. He found oscillations between episodes of intrusion and avoidance to be common after the death and attributed this to incomplete personal schema for understanding death and personal reactions to the event.

Horowitz, Field, & Classen (1993) explained the

motive to repeat the experience as a completion tendency which is based on Freud's (1957) ego-psychological explanation of the compulsive repetition as an attempt of the ego to master and assimilate traumatic experiences by repeated review of the event and its implications. They further explained intrusion as a cognitive attempt to reduce the discrepancy between new information or implications elicited by the event in active memory to pre-existing models of meaning or schemata. The idea was that the new information would eventually be processed to the point where there was no discrepancy and the new information would be assimilated into pre-existing schemata. Traumatic memories and emotions associated with the death were thought to prolong or prohibit assimilation. The ego may not be able to accept the threat of traumatic events and avoidance may be used to defend the person from dreaded states of mind that are too emotional, too painful, too out of control, and too hopeless (Horowitz et al., 1993).

Harris (1991) administered the Impact of Events Scale to 11 healthy adolescents between the ages of 13 and 18 years of age at 1 and 6 months after parental death. All







39

subjects reported moderate to high levels of intrusive and avoidant stress-related symptomatology at the initial assessment with the Impact of Events Scale. Symptoms decreased over the year, but more than half of the subjects continued to experience moderate or high levels of distress

1 year later.

Horowitz et al. (1986) proposed a treatment of stress response syndromes for normal and pathological phases of poststress response. A pattern of phases that individuals tend to progress through in normal responses to serious life events was identified by Horowitz and pathological phasic response to extreme stress was thought to be an intensification of the normal phasic response. He suggests that through a collaborative relationship with the therapist, the patient recounts circumstances of and surrounding the traumatic event. This process involves a reappraisal of the event, the meanings associated with it, and preexisting features of the personality of the patient. The goal is to move the patient from an overcontrolled orientation or undermodulated intrusive reaction to a manageable exploration of the meaning the traumatic event has for the patient.

Transactional Stress Model

According to this theory of stress, psychological stress is a result of a relationship where environmental demands tax or exceed resources of the individual and negative emotions are present based on harm-loss, threat,









and challenge. Stress is considered a relational, processoriented task between the individual and the environment with a goal for cognitive-affective appraisal and coping activities. There are different appraisal processes which serve different functions. Primary appraisal evaluates stressful environmental conditions to determine the nature of stress. Secondary appraisal evaluates which coping options are available and which can be used effectively. Person-environment interactions are constantly evaluated and adjustments are made by coping through this reappraisal process (Lazarus, 1991b).

coping is an activity that regulates emotional

distress and shapes subsequent emotion. It is a planned action that is considered either problem-focused or emotionfocused coping. Problem-focused coping is an attempt to change the actual person-environment relationship by directly acting on the environment or on the self. Emotionfocused coping alters the appraisal component of the mind through avoidance or by changing the meaning of the relationship through denial or distancing, making the distressing emotion moot. If the meaning of an event is changed, the emotion will also be changed (Lazarus, 1991b).

Compas et al. (1988) applied Lazarus' theory of

coping to a study of 130 children from 10 to 14 years of age. They were given an open-ended instrument to assess coping with self-identified recent stressful events. Illness of a friend or family member was frequently listed







41

as a social stressor. The authors found that the number of emotion-focused strategies used with social stressors increased with the age of the child while the use of problem-focused coping was relatively consistent with all age groups.

In developing the instrument, Adolescent Coping

Orientation for Problem Experiences (A-COPE), Patterson and McCubbin (1987) conducted a 3-year longitudinal study designed to examine the influence of family variables on adolescent substance use. Of the subjects, 34% were between the ages of 14 and 18. Gender differences in types of coping were found where females had more frequent use of coping behaviors dlrecmed at developing social support than males. It was found that females report more frequent use of a broader range of coping patterns than males. A-COPE was found to self-educate adolescents about their coping styles and options and to be helpful in teaching adolescents about managing stress.

Adolescent CopinQ Models

Folkman and Lazarus (1980) viewed coping as the

effort to manage specific events or stressors that have been appraised as demanding and as taxing the individual resources. They designed several models to classify coping strategies in adolescents and to explain the relationship between coping and adjustment.

The approach/avoidance coping model conceptualizes

coping in terms of active versus passive strategies in which









active cognitive attempts are used to change the ways of thinking about a problem and behavioral attempts are used to directly resolve the problem. Avoidant strategies include cognitive attempts to deny or minimize threat and behavioral attempts to get away from the problem situation (Billings & Moos, 1981; Lazarus & Folkman, 1984; Roth & Cohen, 1986).

Another model which classifies coping responses

according to their function is the problem/emotion-focused coping model (Folkman & Lazarus, 1980; Menaghan, 1983). Coping strategies are problem-focused attempts to modify the stressor or emotion-focused attempts to regulate emotional states that accompany stress. Compas et al. (1988) found that adolescents who used more problem-focused strategies to cope with interpersonal stressors, like talking things over with others, had fewer emotional and behavioral problems, and those who used more emotion-focused strategies like avoidance or yelling had more emotional and behavioral problems as measured by the Child Behavior Checklist (CBCL) and the Youth Self Report version of the CBCL.

Patterson and McCubbin (1987) derived a scale of

coping responses reported by adolescents which identified 12 direct action and indirect coping response patterns of behavior used when adolescents were faced with difficulties or felt tense. In studies on coping strategies and adolescent substance use, indirect coping through reliance on peer support and ventilating feelings was related to greater substance use, whereas direct coping by seeking







43

greater family support and seeking adult support was related to less substance use (Patterson & McCubbin, 1987; Wills, 1986; Dise-Lewis, 1988; Spirito et al., 1988). In summary, it appears that the effectiveness of adolescent coping varies according to the stage of development, appraisal of the stressor, and coping style.

Person-Based Coping Resources Development

Adolescents are often assumed to be at high risk for psychological stress by developmental stress theorists because they are confronted with many life stressors for the first time without a repertoire of coping responses from which to draw (Konopka, 1980). Spivack and Shure (1982) found that the ability of adolescents to generate a variety of alternative solutions to interpersonal problems is related to behavioral adjustment. Alternatively, youth with emotional and behavioral problems were found to generate fewer alternative solutions than control subjects.

Lazarus and Launier (1978) found that dysfunctional coping is primarily associated with withdrawal, although withdrawal has also been found to be a meaningful way of coping with situations of extreme stress or where direct action is inhibited by external barriers. Age of development, personality, and perception of the stressor have been found to be determinants of the particular styles of coping (Tyszkowa, 1991). Older adolescents have been found to utilize more active problem-solving coping







44

behaviors than younger adolescents who use more emotionbased types of coping (Compas et al., 1988; Patterson & McCubbin, 1991).

Appraisal

Holroyd and Lazarus (1982) define appraisal as the

evaluative process the individual uses to recognize threat or harm and then determine the resources and options available for managing potential or actual harm. Maddi (1981) differentiated between two forms of appraisal that used avoidance and transformational coping. Avoidance coping implies pessimistic cognitive appraisal and evasive action, whereas transformational coping implies more optimistic appraisal and deliberate actions to alter or reduce stressful events. In adult studies of the relationship between coping and cognitive appraisal of stressful situations (Folkman & Lazarus, 1980), it was found that problem-focused coping was used in situations that were appraised as changeable and more emotion-focused coping was used in situations that were appraised as unchangeable realities that must be accepted. Forsythe and Compas (1987) further discovered that psychological symptoms were related to the degree of fit between appraisals of control and the relative amount of problem-focused and emotion-focused coping used. Psychological symptoms occurred when subjects used more emotion-focused coping with events perceived as controllable and more problem-focused coping with events perceived as uncontrollable.







45

In a study with adolescents in grades 6 through 8, Compas et al., (1988) found that adolescents match their coping to fit their appraisal of control over stressful events. Coping strategies were observed with academic stressors such as receiving a poor grade and with interpersonal stressors such as illness of a friend or family member. Females reported using more emotion-focused strategies than males with interpersonal stressors and academic stressors. No gender differences were found between males' and females' use of problem-focused strategies for academic stressors. Adaptive problem-focused coping strategies were found to be more fully developed in older adolescents, whereas emotion-focused coping strategies were not as well-developed in these age groups. Academic stressors were perceived as more controllable than social stressors and more problem-focused alternatives for coping were used with academic stressors. Style

Adolescent coping refers to specific behaviors used to manage demands of a particular situation to reduce or eliminate stress. A pattern of coping responses or a generalized strategy of approaching problems is usually developed that portrays a certain style of coping. Adolescents develop patterns of behaviors to deal with a multiple set of demands that may emanate from their roles as students, children, friends, and life events such as death, moving, divorce, and illness. These styles are developed







46

depending on previous personal experience with similar situations, vicarious experiences associated with observing the success or failure of others, perceptions of personal vulnerability, and social persuasion by parents, peers, and significant others (Patterson & McCubbin 1987; Werner & Smith, 1982). Coping styles emerge during adolescence that will have long-term consequences on coping styles used as adults (Valliant, 1977).

Patterson and McCubbin (1987) found 12 patterns of

coping behaviors during their development of the instrument, Adolescent Coping Orientation for Problem Experiences (ACOPE). The study was conducted with 467 junior and senior high school students from a suburban Midwestern school district. Patterns of behaviors were factor analyzed and grouped into the following categories: ventilating feelings, seeking diversions, developing self-reliance, developing social support, solving family problems, avoiding problems, seeking spiritual support, investing in close friends, seeking professional support, engaging in demanding activity, being humorous, and relaxing. Females were found to have higher mean scores for developing social support, solving family problems, investing in close friends and developing self-reliance. Males had higher mean scores than females on behavior patterns of being humorous. Females were found to use coping behavior more frequently directed at developing social support than males and both genders rated relaxing most frequently as a way to manage tension.







47

Family-Based CopinQ Resources

The family is viewed from a general systems theory (Bertalanffy, 1936) where the members and their relationships bind the system together. Every part of the system is related to its member parts so that change in one will cause change in the total system. Olson and McCubbin (1983) have delineated coping strategies and resources for functional and dysfunctional systems where 16 types of marital and family systems are based on the family dimensions of "cohesion" and "adaptability." This model of family functioning proposes that a balance of these two characteristics is the most functional to marital and family development.

Parent-Adolescent Communication

Clinical literature on childhood bereavement stresses that open communication between parents and children greatly facilitates the mourning process, and children are more likely to achieve a healthy adaptation to parental death when family relationships are characterized by open communication (Bowlby, 1963, 1980a, 1980b; Brice, 1982; Brown, 1988; Caine & Fast, 1966; Furman, R.A., 1964; Furman, E., 1974, 1983; Kliman, 1979; Miller, 1971; Nagera, 1970). Brown (1988) described "openness" as the ability of each family member to stay nonreactive to the emotional intensity in the system and to communicate feelings to others without expecting others to act on them. The ability of the family to remain open, to express thoughts and feelings about the







48

death, and to remain nonreactive to the anxiety of other family members is related to the intensity and duration of stress.

Brown (1988) found that the longer and more intense the family stress is, the greater chance of pathological stress response. In a study of an adolescent son whose father died of cancer, lack of communication in the family caused isolation and anxiety that was expressed in physical symptoms leading to frequent hospitalizations of several family members.

Children's adjustment is less difficult in families

where open expression of anger, guilt, sadness, and loss is encouraged (Vollman, Ganzert, Picker, & Williams, 1971) The general level of communication established in early childhood influences a child's ability to talk about any topic and eventual attitudes toward life and death (Kastenbaum, 1986).

Open communication is of particular importance to

stress management in families because it enables the members to coordinate their efforts in managing demands and helps to reduce ambiguity (Patterson, 1988). McGoldrick (1991) emphasized the importance of clear and open communication about the death and suggested that participation in funerals and graveside services are helpful to family members in integrating loss. Saler and Skolnick (1992) conducted a study of adults whose parent died in childhood and found that their ability to talk freely with the surviving parent







49

about the circumstances around the death, express sorrow to the surviving parent, or ask questions about the deceased parent was protective against adulthood depression. The ability of children to talk about their feelings surrounding death helps them process grief (Moody & Moody, 1991; Segal, 1984; Silverman & Silverman, 1979) and discourages suppression of feelings which might lead to prolonged grieving.

In a study by McNeil (1986), 335 adolescents from 12 to 18 years of age gave their perceptions of family communication patterns. The majority of this sample reported that they disclosed more to their peers before their parents on items relating to feelings about themselves, personal worries, and attitudes and opinions on love, sex, school, and friendships. However, students preferred to disclose their feelings and thoughts about the meaning of life and death and what happens after death to their mothers more than to their peers or fathers. Of the sample, 30% expressed a desire to talk more about deathrelated subjects with their parents than they had in the past.

Goodman (1986) studied 30 adolescents following the death of a parent to observe grief reactions of normal adolescents. Adolescents reported that talking with peers and with others who experienced parental death was helpful in resolving grief but found it difficult to talk to those who had not had this similar experience. Similarly, Cragg







50

and Berman (1990) in a study of 10 adolescents whose parent died of cancer found that adolescents fear being different from their peer groups and that it was difficult for them to turn to peers who had not experienced the loss of a parent. They also found that communication decreased between the surviving parent and adolescents after the death and adolescents were reluctant to discuss their feelings in order to protect the surviving parent from pain.

Problems occur when the surviving parent looks to the children for extraordinary support after the death of a spouse. Children are unable to grieve until their parent has stabilized (Barth, 1989). Family Adaptability

Family adaptability is defined as the ability of the family system to change its power structure of roles and rules in response to situational and developmental stress (Gehring & Feldman, 1988). Barth (1989) interviewed a number of therapists and family members after the death of a parent and found that when problems arise, they are generally noticed 18 months to 2 years after the death. Immediately after the death, the family is preoccupied with soothing one another and not taking care of individual needs. Maladaption may not surface for years and may not emerge until another developmental stage is reached. The family's strengths and resources before parental loss are primary factors in the ability of the family to adjust. Connections with extended family and social support networks







51

before death are also important to adaptation. Their visits and support provide a sense of continuity and safety (Barth, 1989).

Elizur and Kaffman (1983) found that the greatest threat to family adaptability was the surviving parent's inability to express grief, share the child's grief, and incorporate the expression of grief in the family system. Patterson and McCubbin's (1987) Double ABCX Model of Family Adaptation describes development, adaptation, and the influence of demands, resources, meaning, and coping on adolescents. The adolescent, as an individual, is viewed as part of a system together with family and community where reciprocal relationships are formed and the demands of one part of the system are met by the capabilities of another part. There is a "fit" with each system interface-individual-to-family, family-to-community, and individualto-community. Adaptation is achieved from the adolescent's perspective when there is a fit within the family and within the community of peers and in the school system (Patterson & McCubbin, 1987).

Patterson (1988) described a crisis as a state of

disequilibrium when a number of demands exceed the existing capabilities of the family and this imbalance persists. The family reaction during the adaptation phase is to restore balance or normalcy to the family by acquiring new adaptive resources and coping behaviors, reducing demands they must deal with, and changing the way they view their situation.







52

Failure to make these changes may cause an increase in demands and cause the family to be more vulnerable to subsequent stressors.

Davies et al. (1986) studied levels of family

functioning in 11 families where a child had died of cancer from 2 to 9 years earlier. Rigidly maintained roles were found in less functional grieving families. The oldest child was assigned the role of the deceased child by the parents in some families, and in some families the child self-imposed the role. Less functional families were found to be resistant to reorganization and continued to function as if nothing had happened. They were unwilling to make new friends and were unable to mobilize resources. More functional families acknowledged their loss, reorganized and shared responsibilities, and were open, flexible, and adaptive in their reorganization and mobilization of resources.

The Circumplex Model of the family system by Olson

(1989) categorized families according to both their cohesion and adaptability scores. There were 16 subtypes depending on their degree of connection and flexibility. Olson and Stewart (1991) further suggested that there may be a relationship between these subtypes and the effects of stress. "Flexibly connected" families were shown to be more vulnerable to the accumulation of stressful events and "structurally separated" families were more affected by transitional changes. They suggested that life events and







53

transitions are stressful to both types of families and stress is influenced by cohesion and adaptability. Family Cohesion

Family cohesion refers to the emotional closeness that binds family members together and is the degree to which family members are separated from or connected to their family (Olson et al., 1992). Cohesion is generally linearly associated with positive adaptational outcomes in adolescent families. As familial roles and relationships are transformed throughout development, cohesion between parents and children is transformed from relatively high dependency on parents to a balanced connectedness permitting autonomy in adolescence (Noller & Callan, 1986). Power relations also undergo transformations where parental unilateral power decreases and mutual power strategies by adolescents and parents increase (Youniss & Smollar, 1985).

Barth (1989) found that the degree of closeness

children have with parents is a determining factor in how they manage the loss of a parent. In studies of parentally bereaved children, Barth found that a child who is especially close to the surviving parent adjusts more easily to the death of the other parent. Conversely, one who is especially close to the parent who died, and distant from the surviving parent, has more difficulty adjusting.

Elizur and Kaffman (1982) studied groups of bereaved kibbutz and non-kibbutz children at 18-month and 40-month postloss periods. The authors associated several child,









family, and circumstantial variables in nonkibbutz children to suggest poor outcomes of adjustment including poor preloss adjustment, prior marital conflict, poor prior relationships between the child and the deceased parent, poor relations with the surviving parent, and the absence of a surrogate father.

Wolfenstein (1966) theorized that when a parent dies during adolescence, the normal developmental processes of detachment may be avoided by overidealizing the deceased parent. Ambivalence is thought to be created through the overidealizing-detaching conflict and may cause hostility toward the surviving parent. The surviving parent may be blamed for the death and the adolescent may wish the surviving parent had died instead of the deceased.

Alternatively, Wass & Stillion (1988) suggested that

if other family members idealize the deceased parent and the adolescent has been successful in distancing and separating from the deceased, the adolescent may feel cut off and isolated from the other family members and feel misunderstood. In either case, considerable guilt may be generated and behavior problems may become evident. Males have been found to act out by stealing, using drugs, fighting, or demonstrating social withdrawal to escape pain. Whereas, girls were found to sexualize relationships and seek closeness in order to comfort themselves and replace their loss (Osterweis, Solomon, & Green, 1984).







55

Silverman and Silverman (1979) found that it was

easier for bereaved children to accept the loss of a parent if they could maintain a sense of connection to the parent in current life. Silverman and Worden (1992) studied the reactions of a nonclinical group of 125 children from 6 to 17 years of age who had lost a parent. They found that many of the children tried to stay connected to the parent through dreams, talking to or thinking about the deceased parent, claiming that the parent was watching them, visiting the parent's grave, and keeping things that belonged to the parent. The subjects tended to carry on a relationship with the deceased parent similar to the one prevailing before death.

In a retrospective study of college-aged women who

had lost a parent, the women were constantly renegotiating their relationship with their deceased parent. It was thought that by renegotiating a relationship with the deceased parent that it was easier to accept a new reality without the deceased (Silverman et al., 1992).

Conclusion
The death of a parent is believed to be the greatest stressor for an adolescent and family. Researchers have delineated normal and pathological responses to stress and have identified coping mechanisms used to adjust to changes created through loss. Developmental theorists have postulated that cognitive maturity influences coping styles,







56

and life event theorists suggest that life events prepare individuals to cope with subsequent events.

Numerous variables have been identified in the

literature that are important to adolescent adjustment to parental loss, which include cognitive appraisal and decision making, coping strategies, personal resources, parent-adolescent communication, family adaptability, and family cohesion. Very few definitive studies exist, however, relating the adolescent's use of these resources to the stress resulting from parental loss.

This study applies a transactional model of stress to assess the effects of adolescent-environment interactions. Stress response to parental death is measured and the influence of personal and family resources, and coping strategies are assessed. If significant relationships between the levels of stress reported by adolescents and the use of coping and family resources exist, then appropriate interventions might be designed to assist adolescents in developing these personal and family resources.















CHAPTER 3
METHODOLOGY


Statement of Purpose

This study was designed to determine the degree of stress experienced by adolescents as a result of parental death and the relationships among personal and family coping resources and adolescent stress. The family coping resources assessed were family adaptability and cohesion and parent-adolescent communication.

In this chapter the methodology used to address these issues is discussed. The chapter includes a description of the research design, the population and sample, the sampling procedures, the instruments used, the data collection procedures, and the data analysis procedures.

Research Design

A descriptive, correlational design was used with a

qualitative component. The dependent variable in this study was adolescent stress response. Four independent variables used to predict the reported degree of stress were family adaptability, family cohesion, parent-adolescent communication, and types of adolescent coping strategies utilized.

An open-ended questionnaire inviting adolescents to describe the coping strategies they used with the death of







58

the parent and a questionnaire requesting "expert advice" from the adolescent about what would be helpful to a friend who had experienced a similar death comprised the qualitative component of this research study.

Delineation of Relevant Variables Dependent Variable

The dependent variable is stress response.

Subjective distress related to the specific life event of parental death was defined in terms of reported degree of stress. The Impact of Event Scale (IES) (Horowitz et al., 1979) was used to assess the current degree of stress. The scale is comprised of two response sets measuring intrusion and avoidance that when summed, determine the total subjective stress score.

Independent Variables

Family adaptability. Adaptability is the ability of the family system to change its power structure, role relationships, and relationship rules in response to situational and developmental stress (Olson et al., 1992). There is an increase in demands placed on the family after parental death. Death and transitions are stressful to families (Olson & Stewart, 1991) and adaptive resources help reduce the demands family members must deal with after the death of a parent (Patterson, 1988). In this study, family adaptability was measured by the Family Adaptability and Cohesion Scale (FACES II) (Olson et al., 1992).









Family cohesion. Cohesion is defined as the extent

of the emotional bonding that family members have toward one another. The degree of closeness children have with the surviving parent determines the ability of the child to adjust to the loss of a parent (Barth, 1989). In this study, cohesion was measured by the Family Adaptability and Cohesion Scale (FACES II) (Olson et al., 1992).

Parent-Adolescent communication. Parent-adolescent communication was defined as the extent to which the adolescent shares factual and emotional information with the parent (Barnes & Olson, 1985). The ability of family members to express thoughts and feelings about the death of a parent is related to the intensity and duration of stress (Brown, 1988). Open communication between parents and children has been shown to facilitate the grieving process and promote healthy adaptation to parental death (Bowlby, 1980a, 1980b; Brice, 1982; Brown, 1988; Caine & Fast, 1966; Furman, 1983; McGoldrick, 1991; Moody & Moody, 1991; Segal, 1984; Silverman & Silverman, 1979). In this study, parentadolescent communication was measured by the ParentAdolescent Communication Scale (PAC) (Barnes & Olson, 1985).

Adolescent coming. Coping was defined as the ability of the adolescent to regulate emotional distress and actively shape subsequent emotion (Folkman & Lazarus, 1980). Adolescent coping varies according to the stage of development, appraisal of the stressor, and the coping style (Patterson & McCubbin, 1987). Patterns of adolescent







60

behavior portray certain styles when approaching problems and are considered to be problem-focused or emotion-focused coping (Compas, et al., 1988; Patterson & McCubbin, 1991). Coping strategies are used to modify the stressor or to regulate emotional states that accompany stress (Folkman & Lazarus, 1980; Menaghan, 1983). Adolescent coping was assessed by the Adolescent Coping Scale (A-COPE) (Patterson & McCubbin, 1991). Coping styles were identified by the frequencies of twelve descriptive patterns of behavior.

Description of the Population

The population for this study consisted of

adolescents who had lost a parent through death within the past 30 months, were placed in regular classes in school (as opposed to special classes due to emotional or learning disabilities), and were referred by staff members of the Hospice of North Central Florida, school counselors, or by members of the community to the researcher. The Hospice patient population was drawn from an 11-county area of North Central Florida encompassing Alachua, Bradford, Clay, Columbia, Dixie, Gilchrist, Hamilton, Levy, Putnam, Suwannee, and Union counties.

The population in the State of Florida by race in

1990 was 84% White, 14% African American, 1% Hispanic, and 1% other (U.S. Bureau of the Census, 1990). The population of the 11 counties served by the Hospice of North Central Florida in 1990 was 80% White, 18% African American, and 2% other minorities (Bureau of Economic and Business Research,







61

1991). The population of Hospice families treated in 19941995 was 84% White, 14% African American, 1% Hispanic, and 0.3% Other. Table 3-1 shows the percentages by categories for each population.


Table 3-1
Comparative Breakdown by Ethnicity for Population Subsamples


African
Population Subsample White American Hispanic Other


Population of Florida, 1990 84% 14% 1% 1% Population of Hospice of 84% 14% 1% 0.3% North Central Florida, 1990

Research Sample 58.6% 35.7% 3% 1%


Sampling Procedures

Applications to conduct research in public schools were filed in each school district in the 11-county area. The superintendent of schools in each county was mailed a letter (Appendix A) describing the study, together with copies of each instrument used in the study and a copy of a letter to school guidance counselors (Appendix A). A follow-up phone call was made to the superintendent to answer any questions about the project and to determine if approval of the study was granted. The school boards in seven counties agreed to participate and four declined. Upon approval from the superintendent, principals and/or guidance counselors of the middle schools and high schools in each county were contacted and letters were sent to the







62

guidance counselors. The guidance counselors identified students with parental loss and contacted the families to obtain permission to provide names, addresses, and phone numbers for contact by the investigator. Letters were then mailed to surviving parents and legal guardians describing the study (Appendix B) including a letter to the potential adolescent participant (Appendix C). Follow-up phone calls were then placed to the parent/guardians and appointments were scheduled with students who were interested in participating in the study. Arrangements were made to meet the adolescents either at home or at school to complete the questionnaires.

Other referrals were received from Hospice staff members and the coordinator of the Hospice Children's Program. Letters were mailed to the parents along with a letter to the adolescent explaining the research project. Follow-up phone calls were then made to the parent and upon approval of the parent, an appointment was scheduled to meet with the adolescents at their homes or at school.

Ninety-eight letters were mailed to parents and adolescents and 70 agreed to participate in the study. Letters were first mailed on March 16, 1996, and all data was collected by May 11, 1996. There was a 71% return rate on requests to participate. The established minimum of 70 subjects were contacted and no additional procedures for obtaining subjects were instituted. All participants were given five dollars for participating in the study.







63

Upon completion of the project, a letter was mailed to all subjects (Appendix D) thanking them for their participation that included a summary of coping skills reported by participants in this study to be helpful after parental loss.

Sample

The sample consisted of 70 adolescents between the

ages of 12 and 19 (Table 3-2). sixty-nine adolescents were between the ages of 12 and 18, and one 19 year old who was a senior in high school. There were 38 females (54.3%) and 32 males (45.7%) from 7 counties (Alachua [60.0%), Clay [7.14%], Columbia [1.43%], Dixie [1.43%], Levy [7.14%] Putnam [21.43%], and Suwannee [1.43%]).

Ethnicity was represented by 41 Caucasians (58.6%), 25 African Americans (35.7%), 3 Hispanics (4.3%), and 1 other minority (1.4%). There were 26 maternal deaths (37.1%), 40 paternal deaths (57.1%), 1 grandmother (1.4%), and 3 grandfather (4.3%) deaths. Most of the deaths occurred within the past year. Frequencies of deaths included 20 deaths in 1994 (28.6%), 36 in 1995 (51.4%), and 14 within the past 5 months 1996 (20%).

The causes of death were described as follows: 47 due to a long illness (67.1%), 9 due to a sudden illness (12.9),

8 due to an accident (11.4%), 3 murders (4.3%), 1 suicide (1.4%), and 2 other (2.9%).

Frequency distributions by age and county are shown

in Tables 3-2 and 3-3. Frequency distributions by gender of









deceased parent/guardian are shown in Table 3-4. Frequency

distributions by type of death are shown in Table 3-5.



Table 3-2
Frequency Distribution of Participant Age


Total Sample: N = 70
Mean=14.2 SD=1.85
Age N Percent

12 17 24.3 13 11 15.7 14 13 18.6 15 13 18.6
16 8 11.4 17 3 4.3 18 4 5.7 19 1 1.4






Table 3-3
Breakdown by County for Research Sample


County Research Sample Percent

Alachua 42 60.00 Brandford Clay 5 7.14 Columbia 1 1.43 Dixie 1 1.43 Gilchrist Hamilton Levy 5 7.14 Putnam 15 21.43 Suwannee 1 1.43 Union









Table 3-4
Frequency Distributions of Parental Death by Sex


Sex of deceased Percent of Sample


Maternal death 37.1 Paternal death 57.1 Grandmother death 1.4 Grandfather death 4.3


Table 3-5
Frequency Distributions of Parental Death by Type of Death


Cause of Death Research Sample Percent


Long illness 47 67.1% Sudden illness 9 12.9% Accident 8 11.4% Murder 3 4.3% Suicide 1 1.4% Other 2 2.9%



Data Collection

Each adolescent was given a separate collection of

questionnaires, coded in the upper right hand corner with a number (i.e., 101) and was asked not to put their names on the packets. Each participant completed a Demographic Questionnaire, Impact of Event Scale, FACES II, ParentAdolescent Communication Inventory, A-COPE, and Adolescent Questionnaire. There were a total of 128 questions and 2 short-answer questionnaires. Participants completed all questionnaires in 20 to 45 minutes, after which they received five dollars from the investigator for participation in the study. The investigator was present to







66

answer any questions during the sessions and to inform the students of their eligibility to attend bereavement support groups and bereavement camps sponsored by the Hospice of North Central Florida.

Instrumentation

Six instruments were used in this study. Four

standardized instruments were used as follows: (a) the Impact of Events Scale (IES), (b) the Family Adaptability and Cohesion Scales (FACES), (c) the Parent-Adolescent Communication Scale (PACS), and (d) the Adolescent Coping Orientation for Problem Experiences (A-COPE). Two were investigator-developed and included the following: (a) a demographic questionnaire to determine information about the death of the parent and (b) an adolescent questionnaire to determine personal coping skills that adolescents found helpful in dealing with parental death and advice they would give others after similar losses. Measures of DemoQraphic Information

The Demographic Questionnaire (Appendix E) was used to assess the age, gender, ethnicity of the adolescent, county of residence, relation to the deceased, age of the deceased, length of time since the death, and the type of death.

Impact of Event Scale (IES)

The dependent variable was measured by the Impact of Event Scale (IES) developed by Horowitz et al. (1979). It was used to measure subjective distress or internal stress






67

perceived by the adolescent due to the specific life event of the death of a parent.

Subjective distress in response to life events is

theoretically composed of two independent aspects of felt experience: intrusion and avoidance. Intrusion represents the involuntary awareness of ideas, memories, and emotions associated with the event. Avoidance represents the conscious attempt to divert attention away from cognitions and feelings related to the death. The IES includes response sets for both intrusion and avoidance with a sum score for a total measurement of stress.

The scale is comprised of a 15-item self-report measure where 7 items describe episodes of intrusive thoughts and 8 items describe episodes of avoidance. The life event measured--the death of a parent--was listed at the top of the instrument and was the referent for each statement in the scale. The participant was asked to estimate the frequency to which each item described personal experience over the past week on a Likert 4-point scale ranging from "not at all" to "often." A score for each item was obtained by assigning weights of 0 for "not at all," 1 for "rarely," 3 for "sometimes," and 5 for "often." The intrusion subset of questions was represented by items 1, 4, 5, 6, 10, 11, and 14. The avoidance subset of questions is represented by items 2, 3, 7, 8, 9, 12, 13, and 15. An overall sum score presented the degree of current stress in relation to the death of the parent.







68

The test was originally normed on a population of 66 adults who sought psychotherapy as a result of serious life events--half of which were bereaved and the other half who had experienced personal injuries. A cluster analysis was conducted on 20 original items in the test with a primary cluster comprised of clinically derived intrusion items and a second cluster comprised of avoidance items. The test was revised to include 15 items representing the 2 clusters. A split half reliability of the scale was high with r=0.86. Internal consistency of subscales using Cronbach's alpha was 0.78 for intrusion and 0.82 for avoidance. The correlation of 0.42 (p 0.0002) was found between intrusion and avoidance subscale scores, indicating the two subscales were associated but did not measure identical dimensions.

Another population of 35 subjects who were out

patients seeking psychotherapy after the death of a parent were compared to 37 field subject volunteers who experienced the death of a parent. All items were endorsed by both groups. Confirmation of validity of item assignments to intrusion and avoidance subscales was obtained in a factor analysis of the combined data. Subscales were internally consistent in both patient and field groups as well as across time with a Cronbach alpha ranging from 0.79 to 0.92 (Horowitz et al., 1984).

In another study, nonclinical parentally bereaved adolescents were studied over a period of 1 year and evaluations were made at 6 weeks, 7 months, and 13 months






69

following parental death. Stress response symptomatology was measured by the Impact of Event Scale and scores were compared to a study of parentally bereaved adult children. All subjects reported moderate to high levels of intrusive and avoidant stress-related symptomatology initially with decreased symptoms over the year. More than half of the subjects continued to report moderate or high levels of distress 1 year later. Adolescents reported an equal or slightly higher sustained degree of stress compared with parentally bereaved adult patients or control subjects measured by the IES (Harris, 1991). Family Adaptability and Cohesion Scales (FACES-II)

The independent variables of family cohesion and adaptability were measured by FACES II. The scale was originally developed by Olson (1989) as an outgrowth of their Circumplex Model which was a systematic assessment of families and their ability to cope with stress. The instrument was originally developed with the intent to be used with children and those with limited reading ability. During the development of the scale, a 50-item scale was administered to 2,412 individuals in a national survey and on the basis of factor analysis and reliability analysis, the scale was reduced to 30 items. The final scale contained 16 cohesion items and 14 adaptability items.

The cohesion dimension contains 2 items for each of the following eight concepts: emotional bonding, family boundaries, coalitions, time space, friends, decision-






70

making, and interests and recreation. There are four levels of family cohesion ranging from extreme low cohesion to extreme high cohesion and families are characterized as disengaged, separated, connected, and enmeshed. The balanced levels of family cohesion are known as separated and connected.

The adaptability dimension contains 2 to 3 items for the following 6 concepts: assertiveness, leadership, discipline, negotiations, roles, and rules. The four levels of family adaptability range from extreme low adaptability or change to extreme high adaptability or change. The levels are characterized as rigid, structured, flexible, and chaotic. The balanced levels are known as flexible and structured. Balanced levels are generally viewed as being characteristic of healthy family functioning and extreme levels are seen as more problematic for families over time.

Internal consistency reliability for FACES II was initially derived from a study with a sample of 124 university and high school students with an average age of 19.2 years. Reliability was .87 for cohesion, .78 for adaptability, and .90 for the total scale. Test-retest reliability measures were taken from the-50 item version and the time lapse between the first and second administration for the test was 4 to 5 weeks. The Pearson correlation for the scale was .84 (.83 for cohesion and .80 for adaptability).







71

The FACES II was administered to each participant. The participant was asked to read the statements and determine how frequently, on a scale from 1 "almost never" to 5 "almost always" the described behavior occurs in the family. Linear scoring suggests that high scores on adaptability and cohesion dimensions indicate "very connected" and "very flexible" measures. After cohesion and adaptability scores were obtained, corresponding scores for family type were interpreted through the linear scoring and interpretation chart by Olson et al. (1992). Family types were described as balanced, moderately balanced, mid-range, and extreme.

Parent-Adolescent Communication Inventory (PACI)

The independent variable, parent-adolescent

communication, was measured by the Parent-Adolescent Communication Inventory (PACI) developed by Barnes and Olson (1985). PACI is a self-report instrument designed to measure both positive and negative aspects of communication as well as aspects of the content and process of parentadolescent interactions. The scale was initially normed on a sample of 433 students with the majority in the late adolescent age range of 16 to 20 years of age. There were 127 high school students and 306 college students in the study.

Thirty-five items were selected from a review of the literature on parent-adolescent communication and other items were generated to measure different aspects of parent-







72

teen interaction. Items were selected to measure both process and content of communication and provided a readable scale for youth as young as 12 years of age. Data from the pilot study were analyzed using factor analysis methods and

3 subscales were identified as open family communication, problems in family communication, and selective family communication. The scale was reduced to 20 items based on factor analysis. The scale was then administered to 124 subjects with a mean age of 19.2 years to determine the test-retest reliability with intervals of 4 to 5 weeks.

Subsequent studies were conducted with adolescents and a final scale was then developed to consider both adolescent and parent views. The scale was designed to measure both positive and negative aspects of communication as well as aspects of the content and process of the parentadolescent interactions. The 20-item Likert-type scale consists of 2 subscales. The first subscale, "open family communication," measures positive aspects of parentadolescent communication with a focus on free flowing exchange of information, lack of constraint, and degree of satisfaction in interactions. The second subscale, "problems in family communication," focuses on negative aspects such as hesitancy to share, negative styles of interaction, and caution in selectivity of what is shared.

The final scale was administered to a larger study with a total sample of 1,842 subjects. Sample I contained 925 and Sample II contained 916 subjects. The alpha







73

reliabilities were .87 for open family communication, .78 for problems in family communication, and .87 for the total score indicating that the two subscales and the total scale are reliable. Although adolescent responses regarding each parent were similar, the intergroup differences were significant enough to warrant the need for reporting different norms. Norms are reported for four different subdivisions: 1) fathers reporting an interaction with teens, 2) mothers' reports, 3) adolescents' reports regarding their mothers, and 4) adolescents' reports regarding their fathers.

Participants were given the 20 statement inventory and asked to rate responses on a Likert-style 5-point response range from "strongly disagree" to "strongly agree" for a total sum score. A total sum score is used to reduce response bias. Items from the subscales are intermingled and it is necessary to distinguish items from the subscales for scoring. Items from the first subscale, which measure more positive aspects of parent-adolescent communication, are scored by adding the response choices in the Likert-type scale (1=strongly disagree; 2=moderately disagree; 3=neither agree nor disagree; 4=moderately agree; 5=strongly agree). Items from the second subscale, denoting problems in family communication, are reversed in value. These items must be flipped in point value prior to summing the response values of the second scale with the first subscale for a sum total score of positive parent-adolescent communication.







74

Adolescent CopinQ Orientation for Problem Experiences

The independent variable, adolescent coping behavior, was measured by the adolescent coping orientation for problem experiences (A-COPE) developed by Patterson and McCubbin (1991). Three adolescent population samples were used by the developers of the instrument. The first sample consisted of 30 10th, 11th, and 12th grade males and females. They were asked to describe what they do to manage hardships and to relieve discomfort for a) the most difficult stressor event personally experienced, b) the most difficult stressor event experienced by another family member, and c) difficult life changes and strains experienced in general. Ninety-five coping behavior items were generated which reflected the adolescent's developmental tasks of developing an identity and differentiating from the family while simultaneously maintaining appropriate linkages to it. Coping behaviors represented three primary functions of coping: a) coping by direct action to modify the situation by reducing demands and/or increasing resources, b) coping by altering the meaning of experiences through perception and appraisal, and c) coping by managing the tension or stress experienced.

Sample 2 consisted of 467 junior and senior high

school students including 185 males and 241 females. The average age was 15.6 years. The 95 coping behavior items were presented and adolescents were asked to record how often they used each behavior when they faced difficulties







75

or felt tense. The responses were factor analyzed to determine the underlying structure of A-COPE.

Sample 3 consisted of 709 adolescents and their parents who were part of a 3-year longitudinal study designed to examine the influence of family variables on the use of substances by adolescents. Ages of adolescents ranged from 11 to 18 years old. The adolescents were asked to complete the A-COPE and an annual substance abuse questionnaire.

A hierarchical approach to coping was then developed with A-COPE. Two general levels of abstraction were defined: (a) coping behaviors were operationally defined through 95 specific items and (b) specific coping patterns which were combinations of coping behaviors. Responses of 467 adolescents from sample 2 identified 95 coping behavior items of which 27 items were deleted. Repeated factor analyses resulted in a final set of 54 coping behavior items with factor loadings above .40 on twelve factors with eigenvalues of 1.0 or greater. The 12 factors accounted for 60.1 percent of the variance of the original correlation matrix. The factor structure of the coping behaviors in ACOPE with respective Cronbach alpha reliabilities on 12 factors are presented in the following twelve coping patterns: ventilating feelings (alpha=.75), seeking diversions (alpha=.75), developing self-reliance and optimism (alpha=.69), developing social support (alpha=.75), solving family problems (alpha=.71), avoiding problems









(alpha=.71), seeking spiritual support (alpha=.72), investing in close friends (alpha=.76), seeking professional support (alpha=.50), engaging in demanding activity (alpha=.67), being humorous (alpha=.72), and relaxing (alpha=.60).

Concurrent validity was examined in relationship to

substance use in adolescents in sample 3. Substance use was reported to have positive associations with coping patterns of ventilating feelings for males (r=.13 to .17), females (r=.11 to .17), and developing social support for males (r=.16 to .25), females (r=.13 to .18). Positive associations between friend support and substance use suggest that the socializing role of peers may influence adolescents to use substances. Coping patterns that had negative associations with substance use were solving family problems for males (r=-.13), females (r=-.10 to -.21) and engaging in demanding activity for males (r=-.12), females (r=-.13 to -.18). Seeking spiritual support for females was negatively associated with substance use (r=-.11 to -.21). Other examples of coping patterns and decrease of substance use reflect the reliability and validity of A-COPE as an instrument for measuring adolescent coping with life problems.

Coping behaviors cluster into coping patterns which appear to have a discrete function. For example, "ventilating feelings" is a coping pattern and the following behaviors are associated: get angry and yell at people,









blame others for what's going wrong, say mean things to people, swear, let off steam by complaining to friends, and let off steam by complaining to family members. Behaviors associated with ventilating feelings and avoiding problems are considered undesirable or negative, whereas the remaining 10 patterns are considered transformational coping or positive types of coping behaviors. Five of the coping patterns (developing social support, solving family problems, seeking spiritual support, investing in close friendships and seeking professional support) involve talking to other people to discover solutions to problems and to increase social support and are considered to be "direct action" or problem-focused coping. However, these behaviors may also help reduce tension and may indirectly lead to altered meanings. The coping process appears to be multidimensional and behaviors may serve more than one function (solve problems and manage emotions). There have been reports of gender differences in patterns of coping behaviors where females used coping behaviors directed at developing social support more frequently than males, and males used behaviors directed at being humorous more frequently than females (Patterson, 1988).

Functions of coping (Lazarus, 1966; Menaghan, 1983; Moos & Billings, 1982; Pearlin & Schooler, 1978) are described as (a) direct actions to reduce demands or increase resources through problem-focused coping, (b) altering the meaning through appraisal, and (c) managing







78

tension through emotion-focused coping. Moriarty and Toussieng (1976) suggest that coping behavior may serve more than one function and that coping involves both a response to demands of external situations and a response to one's feelings about the situation.

An adolescent coping score for A-COPE is simply a

total sum score which represents positive coping patterns of behavior. The 54-item Likert-style instrument is administered and the participant is asked to circle responses from 1 "never" to 5 "most of the time." The coping score is determined by summing the scores for each of the items. However, 9 items are scored with reversed values to ensure that all items are weighted in the same positive direction for both the analysis and interpretation of results.

Adolescent Questionnaire (ApDendix F)

The adolescent questionnaire is a two-part

questionnaire designed by the investigator as a qualitative component of the study. It was formulated to (a) identify coping behaviors used by the adolescents that were helpful after the death of their parents and coping that is used presently and (b) to solicit "expert advice" from the adolescents about what they would tell their best friends if they experienced a similar situation with the death of one of their parents. The first part of the questionnaire was designed to allow the adolescent to discuss coping experiences freely without a predetermined list of coping







79

behaviors. The second part of the questionnaire was designed to allow the adolescents to reflect upon their losses and again decide what was helpful to them, but from a different perspective. The intent was to put the adolescents in an "expert role" such that the advice they might give a friend may be different from what they experienced but important to the grieving process.

Hypotheses

The following null hypotheses were evaluated in this study:

1. After controlling for age and length of time

since death, there is no relationship between adolescents' use of positive coping strategies and their reported degree of stress following the death of a parent.

2. After controlling for age and length of time

since death, there is no relationship between adolescents' perceptions of the extent of positive parent-adolescent communication and their reported degree of stress following the death of a parent.

3. After controlling for age and length of time

since death, there is no relationship between adolescents' perceptions of the degree of family cohesion and their reported degree of stress following the death of a parent.

4. After controlling for age and length of time

since death, there is no relationship between adolescents' perceptions of the degree of family adaptability and their reported degree of stress following the death of a parent.









Data Analysis

Analyses of data were conducted using multiple

regression to determine the relationships between adolescent stress and personal and family coping resources. Other regression analyses were used to assess the relationships between overall coping strategy and "coping patterns" of behavior.

Means, standard deviations, and score ranges were computed to describe the sample in terms of intensity of stress experienced, types of coping strategies and patterns of coping styles utilized, degree of family cohesion, degree of family adaptability, and the degree of parent-adolescent communication.

Adolescent questionnaires were analyzed and all

coping mechanisms used by participants were categorized and tabulated regarding frequencies of coping skills considered helpful to adolescents after the death of a parent. Similar analyses were conducted regarding the advice adolescents would offer a friend who had lost a parent. Specific data and analytic procedures and results are described in Chapter 4.















CHAPTER 4
DATA ANALYSIS AND RESULTS


This study was designed to examine the degree of stress in adolescents as a result of parental death. A second purpose of the study was to explore the relationship between the use of personal coping strategies and adolescent stress. The third purpose was to explore the mediating effects of family adaptability, family cohesion, and parentadolescent communication on adolescent stress.

In addition, a qualitative analysis was used to

assess personal coping mechanisms reported by adolescents after the death of the parent and coping skills that continue to be helpful in managing their grief. Adolescents also were asked for the advice they would give their "best friend" after a similar experience of parental death.

In this chapter, the results of the study will be presented as they pertain to each of the four null hypotheses. Next, the results of the qualitative analysis will be reported in terms of frequencies of coping mechanisms described and coping strategies recommended for friends.

Four null hypotheses were evaluated in this study. Analyses of data were conducted using multiple regression with the SAS General Linear Model (GLM) to determine the







82

relationships between the outcome measure of stress and the independent measures of coping, parent-adolescent communication, family cohesion, and family adaptability. Further regression analyses were conducted to determine the relationships between the 12 patterns of coping behaviors and gender of adolescent, relationship of deceased, age of deceased, length of time since death, communication, cohesion, and adaptability. Means, standard deviations, and score ranges were computed to describe the sample. For purposes of determining levels of statistical significance, the Type I error rate of .05 was established as the decision rule for all statistical tests.

Hypothesis One

The first null hypothesis addressed the relationship between adolescent use of positive coping strategies and subjective stress. Means and standard deviations of the total stress scale as well as individual subscales of intrusion and avoidance are reported in Table 4-1. In addition, the sample mean and standard deviation are reported for coping, communication, cohesion, and adaptability measures. As noted, participants scored a total mean score of 37.36 (SD=l0.34, range 16-59) on the Impact of Event Scale. Scores are similar to a parentally bereaved adolescent sample (mean 39) reported by Harris (1991) and a sample of parentally bereaved neurotic adults (mean 39.5 (SD=17.12, range 0-69) reported by Horowitz et al. (1979).









Table 4-1
Means, Standard Deviations, and Ranges of the Dependent Variable, Stress, and Independent Variables: Coping, Communication, Cohesion, and Adaptability


Variable Mean SD Range


STRESS 37.36 10.34 16-59
Intrusion 19.16 5.57 7-28 Avoidance 18.29 6.20 8-32 COPING 168.05 26.37 101-225 COMMUNICATION 64.79 15.16 20-91 COHESION 52.73 12.29 24-74 ADAPTABILITY 43.34 9.38 14-60



The subscale of intrusion (Table 4-1) had a mean

score of 19.16 (SD=5.57, range 7-28) compared to a sample of neurotic adults who had a higher mean of 21.4 (SD=9.6, range 0-35). The avoidance subscale (Table 4-1) of subjects in this study had a mean of 18.29 (SD=6.20, range 8-32) and was similar to the sample of neurotic adults with a mean of 18.2 (SD=I0.8, range 0-38).

Measurements of coping on the A-COPE (Table 4-1)

produced a mean score of 168.05 (SD=26.37, range 101-225). This mean is similar to normed scores for Black adolescents (mean of 169.43, SD=25.23) and White adolescents (mean 167.86, SD=24.45) from single parent families (Patterson & McCubbin, 1991).

Null hypothesis one states that after controlling for age and length of time since death, there is no relationship







84

between adolescents' use of positive coping strategies and their reported degree of stress following the death of a parent.

A series of three multiple regression analyses were conducted. In the first multiple regression analysis in which the total stress score served as the dependent variable, there was no significant relationship between the independent variable, coping, and the dependent variable, the total stress score. Results of the analysis are shown in Table 4-2.

In a second multiple regression analysis, in which the intrusion subscale score served as the dependent variable, no significant relationship was found between the variables (R-Square=0.09, p=.77). In the third multiple regression analysis, in which the avoidance subscale score served as the dependent variable, no significance was found among variables (R-Square=.06, p=.89). The model failed to achieve significance for the independent variable of coping (F=.52, p=.47) (Table 4-3). Therefore, no statistical evidence existed to reject the null hypothesis. Analyses of subscales are shown in Tables 4-3 and 4-4.

Null Hypothesis Two

The second null hypothesis addressed the relationship between positive parent-adolescent communication and subjective stress. A mean score of 64.79 was computed for









Table 4-2
Hierarchial Regression Analysis of the Relationships Among Age, Length of Time Since Death, Communication, Cohesion, Adaptability with Total Stress Score as the Outcome Variable



Total Sample N = 70
Variables df Type III SS F-value p-value


Age 1 7.4749 0.07 0.78 Length 1 121.5754 1.19 0.28 Cope 1 53.3848 0.52 0.47 Comm 1 24.3430 0.24 0.62 Cohes 1 99.3632 0.97 0.32 Adapt 1 32.4367 0.32 0.57



Variables Est SE t-value p-value


Intercept 38.25 14.53 2.63 0.01 Age 0.22 0.81 0.27 0.78 Length -2.24 2.05 -1.09 0.28 Cope 0.04 0.05 0.72 0.47 Comm -0.07 0.14 -0.49 0.62 Cohes -0.18 0.19 -0.99 0.32 Adapt 0.14 0.26 0.56 0.57

R-Square=0.09


*p <.05









Table 4-3
Hierarchial Regression Analysis of the Relationships Among Age, Length of Time Since Death, Communication, Cohesion, Adaptability with Intrusion Subscale Score as a Measure of Stress



Variables df Type III SS F-value p-value


Age 1 0.0127 0.00 0.98 Length 1 39.3602 1.32 0.25 Cope 1 58.2020 1.96 0.16 Comm 1 5.3433 0.18 0.67 Cohes 1 14.3611 0.48 0.49 Adapt 1 2.5583 0.09 0.77



Variables Est SE t-value p-value


Intercept 17.28 7.80 2.21 0.03 Age 0.00 0.43 0.02 0.98 Length -1.24 1.08 -1.15 0.25 Cope 0.04 0.03 1.40 0.16 Comm -0.03 0.07 -0.42 0.67 Cohes -0.07 0.10 -0.69 0.49 Adapt 0.04 0.14 0.29 0.77

R-Square=0.09


* p <.05









Table 4-4
Hierarchial Regression Analysis of the Relationships Among Age, Length of Time Since Death, Communication, Cohesion, Adaptability with Avoidance Subscale Score as a Measure of Stress



Variables df Type III SS F-value p-value


Age 1 2.9581 0.07 0.78 Length 1 44.5796 1.11 0.29 Cope 1 0.0062 0.00 0.99 Comm 1 3.3910 0.08 0.77 Cohes 1 42.5985 1.06 0.30 Adapt 1 14.8818 0.37 0.54


Variables Est SE t-value p-value


Intercept 22.24 8.85 2.51 0.01 Age 0.13 0.48 0.27 0.78 Length -1.32 1.26 -1.05 0.29 Cope -0.00 0.03 -0.01 0.99 Comm -0.02 0.09 -0.29 0.77 Cohes -0.12 0.12 -1.03 0.30 Adapt 0.10 0.16 0.61 0.54 R-Square=0.06


* p <.05


this sample on the Parent-Adolescent Communication Inventory indicating an average sense of positive communication with surviving parents. This mean score is similar to both reported norms by Barnes and Olson (1985) for motheradolescent communication (mean 66.56, SD=I2.10, range 20100) and for father-adolescent communication (mean 63.74, SD=12.02, range 20-100). The aspects of parent-adolescent communication assessed by the instrument were the amount of openness, the extent of problems or barriers of family







88

communication, and the degree to which adolescents are selective in their discussion with parents.

Null hypothesis two states that after controlling for age and length of time since death, there is no relationship between adolescents' perception of positive parentadolescent communication and their reported degree of stress following the death of a parent.

The results of multiple regression analysis indicate that there was no significant relationship between parentadolescent communication and reported degree of stress. The model failed to denote significance between parentadolescent communication and stress (F=.24, p=.62) (Table 42). Therefore, the null hypothesis was not rejected.

Null Hypothesis Three

The third null hypothesis addressed the relationship between family cohesion and subjective stress. The mean score for family cohesion was 52.73 (SD=12.29, range 20-91). The linear scoring and interpretation procedure (Olson et al., 1992) indicates that the mean score for this sample denotes the average family as "separated" on the cohesion subscale of the FACES II. While this score falls within the balanced range of family types, it is slightly lower than an adolescent normed sample score of 56 reported by Olson and McCubbin (1983). This suggests that the score is consistent with studies of adolescents who typically report lower family cohesion because of their developmental stage and increased autonomy.







89

Null hypothesis three states that after controlling for age and length of time since death, there is no relationship in adolescents' perceptions of the degree of family cohesion and their reported level of stress following the death of a parent.

The results of multiple regression analysis indicate that there was no significant relationship between family cohesion and the reported degree of stress (F=.97, p=.32). The null hypothesis was not rejected. Results are shown in Table 4-2.

Null Hypothesis Four

The fourth null hypothesis addressed the relationship between family adaptability and subjective stress. A mean score for adaptability in this study (43.34, SD=9.38, range 14-60) was slightly lower than a normed sample of adolescents with a mean score of 46 reported by Olson and McCubbin (1983). The linear scoring and interpretation scale (Olson et al., 1992) was used to characterize the family as "structured" based on scores of the FACES II. The combined subscores for cohesion (52.73, "separated") and adaptability (43.34, "structured") indicated that the "structurally separated" scores produced an average family type description of "mid-range" families. Olson and Stewart (1991) reported that "structurally separated" or mid-range family types are not as vulnerable to stressful life events as much as extreme family types.







90

Null hypothesis four states that after controlling for age and length of time since death, there is no relationship in adolescents' perceptions of the degree of family adaptability and their reported degree of stress following the death of a parent.

The results of multiple regression analysis indicate that there were no significant effects for family adaptability and the reported degree of stress (F=.32, p=.57). Therefore, the null hypothesis was not rejected. Analyses are shown in Table 4-2.

Adolescent Use of CopinQ Strategies

While stress was the major focus in this study, an

understanding of how participants utilized particular coping styles of behavior in managing stress was also important. A follow-up analysis was conducted to assess the frequencies with which certain coping styles were used by the sample and to assess the relationships between the twelve patterns of coping behaviors and the independent variables.

Within the coping scale, there are distinct clusters of coping behaviors that represent certain styles of coping. These behaviors are classified on the basis of function, and different patterns may serve more than one function. Means and standard deviations for coping styles used by this sample are reported in Table 4-5.




Full Text

PAGE 1

THE DIFFERENTIAL IMPACT OF PARENTAL DEATH ON ADOLESCENT STRESS AS DETERMINED BY INDIVIDUAL AND FAMILY COPING RESOURCES By LUCIA PATAT YOUNG 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 1996 UNIVERSITY OF FLORIDA LIBRARIES

PAGE 2

Dedicated to my husband, Dr. Michael Young, son, David Young, and my daughter, Allison Young for the support and encouragement they gave me throughout this endeavor. I dedicate this also to the memory of my dear father, Leon Patat, and to my mother, Amelia Patat.

PAGE 3

ACKNOWLEDGMENTS I wish to thank the following people for their help in completing this dissertation. I am especially grateful to Dr. David Miller for his encouragement, his kindness, and his depth of knowledge. I wish to thank the other members of my supervisory committee — Dr. James Pitts for his sense of humor and helpful suggestions. Dr. Robert Ziller for the fun, theoretical discussions, and especially my chairperson, Dr. Ellen Amatea, for her leadership. I am grateful to all of the participants in my study who "talked with me" about their thoughts and feelings regarding the death of their parents, and I am hopeful that their recommendations to others might be shared through the publication of this study. I am deeply grateful to my family for their understanding, support, encouragement, and love throughout this entire process. iii

PAGE 4

TABLE OF CONTENTS page LIST OF TABLES V ABSTRACT vii CHAPTERS 1 INTRODUCTION 1 Scope of the Problem 5 Theoretical Framework 12 Need for the Study 16 Purpose of the Study 17 Research Questions 18 Significance of the Study 19 Definitions 19 Organization of the Study 21 2 LITERATURE REVIEW 23 Introduction 23 Impact of Parental Death on the Adolescent 23 Normal Stress Response to Parental Death ... 26 Pathological Stress Response to Parental Death 28 Theoretical Models of Childhood Bereavement .... 32 Stress Theories 35 Life Events Stress Model 36 Transactional Stress Model 39 Adolescent Coping Models 41 Person-Based Coping Resources 43 Development 43 Appraisal 44 Style ! ! 45 Family-Based Coping Resources 47 Parent-Adolescent Communication 47 Family Adaptability 50 Family Cohesion [53 Conclusion ! ! ! ! 55 3 METHODOLOGY 57 Statement of Purpose 57 Research Design ' 57 iv

PAGE 5

Delineation of Relevant Variables 58 Dependent Variable 58 Independent Variables 58 Description of the Population 60 Sampling Procedures 61 Sample 63 Data Collection 65 Instrumentation 66 Measures of Demographic Information 66 Impact of Event Scale (lES) 66 Family Adaptability and Cohesion Scales (FACES-II) 69 Parent-Adolescent Communication Inventory (PACI) 71 Adolescent Coping Orientation for Problem Experiences 74 Adolescent Questionnaire (Appendix F) 78 Hypotheses 79 Data Analysis 80 4 DATA ANALYSIS AND RESULTS 81 Hypothesis One 82 Null Hypothesis Two 84 Null Hypothesis Three 88 Null Hypothesis Four 89 Adolescent Use of Coping Strategies 90 Adolescent Questionnaire 98 Summary 100 5 DISCUSSION 101 Description of Sample lOi Question One: Relationship Between Stress and Coping 102 Question Two: The Relationship Between Stress and Parent-Adolescent Communication 104 Question Three: Relationship Between Stress and Cohesion 105 Question Four: The Relationship Between Stress and Adaptability 107 Adolescent Questionnaire: Coping Behaviors and Advice 108 Limitations of the Study .* 109 Implications: Counseling, Future Research, and Theory 2.11 Summary ! ! ! * * 114 APPENDICES A LETTER TO SCHOOL COUNSELORS 116 B PARENTAL CONSENT LETTER 11-7 V

PAGE 6

C LETTER TO ADOLESCENT 119 D CHILD ASSENT SCRIPT 120 E DEMOGRAPHIC QUESTIONNAIRE 121 F ADOLESCENT QUESTIONNAIRE 122 G LETTER TO PARTICIPANTS 123 REFERENCES 124 BIOGRAPHICAL SKETCH 137 vi

PAGE 7

LIST OF TABLES page 3-1 Comparative Breakdown by Ethnicity for Population Subsamples 61 3-2 Frequency Distribution of Participant Age .... 64 3-3 Breakdown by County for Research Sample 64 3-4 Frequency Distributions of Parental Death by Sex 65 35 Frequency Distributions of Parental Death by Type of Death 65 41 Means, Standard Deviations, and Ranges of the Dependent Variable, Stress, and Independent Variables: Coping, Communication, Cohesion, and Adaptability 83 4-2 Hierarchial Regression Analysis of the Relationships Among Age, Length of Time Since Death, Communication, Cohesion, Adaptability with Total Stress Score as the Outcome Variable 4-3 Hierarchial Regression Analysis of the Relationships Among Age, Length of Time Since Death, Communication, Cohesion, Adaptability with Intrusion Subscale Score as a Measure of Stress 4-4 Hierarchial Regression Analysis of the Relationships Among Age, Length of Time Since Death, Communication, Cohesion, Adaptability with Avoidance Subscale Score as a Measure of Stress 1-5 Means, Standard Deviations, and Ranges of the Twelve Patterns of Coping Behavior Were Computed to Assess Coping Styles Used by Subjects in this Study vii

PAGE 8

4-6 Beta Weights for Relationships Between Independent Variables and Coping by Engaging in Demanding Activity 93 4-7 Beta Weights for Relationships Between Independent Variables and Coping by Seeking Professional Support 93 4-8 Beta Weights for Relationships Between Independent Variables and Coping by Avoiding Problems 95 4-9 Beta Weights for Relationships Between Independent Variables and Coping by Solving Family Problems 95 4-10 Beta Weights for Relationships Between Independent Variables and Coping by Developing Social Support 97 4-11 Beta Weights for Relationships Between Independent Variables and Coping by Seeking Diversions 98 viii

PAGE 9

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 THE DIFFERENTIAL IMPACT OF PARENTAL DEATH ON ADOLESCENT STRESS AS DETERMINED BY INDIVIDUAL AND FAMILY COPING RESOURCES By Lucia Patat Young August 1996 Chairman: Dr. Ellen S. Amatea Major Department: Counselor Education The death of a parent is considered to be the most stressful life event for the adolescent. Adjustment to the new demands and life changes resulting from parental loss are accomplished through the use of both personal and family coping resources. Significant relationships were hypothesized between the degree of stress and the use of certain coping strategies, the levels of family cohesion and adaptability, and the style of parent-adolescent communication utilized with the surviving parent. In this study, individual coping theory, family stress theory, and transactional stress theory provide a foundation for assessing adolescent stress. A sample of 70 parentally bereaved adolescents were assessed in terms of the above mentioned variables through a series of multiple regression analyses. Outcome assessments ix

PAGE 10

were represented by a total stress score and subscale scores of intrusion and avoidance. No significant associations were found between the variables and the stress scores. However, the degree of stress reported for the sample was found to be moderate to high. Measures of family cohesion and adaptability indicated that the families of these adolescents were perceived to be of a "mid-range" type. Parent-adolescent communication was reported to be open and positive. A post hoc analysis of the relationships between coping strategies and family cohesion, parent-adolescent communication, length of time since death, gender of adolescent, and gender of deceased revealed some interesting patterns of coping styles. The most frequently reported coping pattern of behavior in dealing with the death was "avoidance," although the majority of coping styles used were direct or problem-solving methods. In addition to standardized assessments, adolescents were asked to discuss coping strategies that were helpful to them in dealing with the death and to discuss the advice they would give to a friend who lost a parent. Analyses indicated that the majority of participants utilized the help of social and family support and advised friends specifically "to talk" about their experiences with a friend or family member.

PAGE 11

CHAPTER 1 INTRODUCTION Adolescents report the death of a parent to be the most stressful event in their lives (Coddington, 1972; Crook & Eliot, 1980; Dise-Lewis, 1988; McNeil, 1989; Raphael, 1983; Siegel, Mesagno, & Christ, 1990). The loss is a harsh separation that is beyond their control, and they struggle not only to reorient themselves to their environment but also to redefine their relationships with their deceased parents. Moreover, the loss of a parent may generate other stressors for the adolescent such as increased family responsibilities, problematic communication with surviving parents, physical or mental illness in surviving parents, social isolation from friends, economic insecurity, and premature separation from parents (Vida & Grizenko, 1989) . The challenge of adapting to these painful circumstances can be an overwhelming one. Most adolescents report initial grief reactions such as shock, numbness, and disbelief (Raphael, 1983; Silverman & Worden, 1992). They may demonstrate feelings of sadness, guilt, anger, depression, disorientation, and insecurity (Unpublished manuscript, Gapes, C. , 1982) and exhibit behaviors of crying, insomnia, learning difficulties, withdrawal, appetite problems, and temper tantrums (Gray, 1987; Harris, 1

PAGE 12

2 1991; Silverman & Worden, 1992; Vida & Grizenko, 1989). Some adolescents, however, demonstrate severe symptoms over a long period of time. For example, unresolved, suppressed, and chronic grief response may be acted out in behaviors of school delinquency, increased alcohol and drug intake, phobias, suicide ideation, depression, antisocial behaviors, academic failure, stealing, fighting, reckless driving, unwanted pregnancy, and hostility toward the surviving parent (Freudenberger & Gallagher, 1995; Raphael, 1982, 1983) . Compulsive caregiving of the surviving parent is another symptom often demonstrated in which the adolescent suppresses personal grief in an attempt to protect the surviving parent from pain. Many times the adolescent is pressured to take on the role of the deceased parent to regulate emotional distress in the family (Bowlby, 1980b; Cheifetz, Stavrakakis, & Lester, 1989). Some adolescents develop symptoms of psychopathology during the first year of parental bereavement (Kaffman & Elizur, 1973; Kliman, 1979; Raphael, 1982) while others demonstrate a "sleeper effect" reporting symptoms several years after the loss (Harris, 1991; Vida & Grizenko, 1989). A common reaction to parental death is to regress and demonstrate behaviors characteristic of a younger aged child. To guard against regression, some adolescents demonstrate an inability to cry or express emotion for fear of becoming overwhelmed with grief and feeling like a helpless child (Adams-Greenly & Moynihan, 1983). Many

PAGE 13

3 younger adolescents appear preoccupied with the death. In contrast, older teens may use denial and avoidance in dealing with the emotional distress (Harris, 1991) . What distinguishes adolescents who do adapt over time to this painful loss from those who do not? Although researchers have shown increased interest in studying adolescent response to parental death, few efforts have been made to explore the differences between adolescents who have greater difficulty adapting to changes created by parental death than those who are able to adapt more readily. One promising line of inquiry involves examining the personal and family resources adolescents bring to bear in coping with their loss. Jackson and Bosma (1990) propose that there is a strong relationship between coping and selfconcept in adolescents and suggest that personal coping resources such as internal locus of control, high selfesteem, and increased age and cognitive ability help protect the adolescent in stressful situations. They suggest that if the adolescent is able to establish a positive identity, a self-concept as adults will be forged that is capable of coping successfully with stresses arising in adulthood (Jackson & Bosma, 1990). Silverman, Nickman, and Worden's (1992) "child bereavement study" suggests that after parental death, there is a process of adaptation where the self-concept is reshaped to accommodate a new relationship with the deceased by incorporating the loss. A new reality is constructed that maintains a connection with the parent.

PAGE 14

4 This cognitive-emotional approach is described as an ongoing process throughout development that facilitates personal coping and accommodation to life changes. Researchers have found numerous aspects of family life which influence the coping behaviors of bereaved adolescents. These include family communication, cohesion, and adaptability (Adams-Greenly & Moynihan, 1983; Buehler, 1990; Cragg & Herman, 1990; Crosby & Jose, 1983; Gray, 1989; McGoldrick, 1991; Moody & Moody, 1991; Noller, Seth-Smith, Bouma, & Schweitzer, 1992; Patterson, 1988; Patterson & McCubbin, 1991; Saler & Skolnick, 1992; Silverman & Silverman, 1979; Stern, Van Slyck, & Newland, 1992; Siegel et al., 1990; Walker, 1985). The presence of a supportive family environment, including parental warmth, cohesiveness, and organization, has been found to provide positive models for identification (Compas, 1987). Open communication with surviving parents about a parent's death facilitates the ability to mourn (Siegel & Gorey, 1994). In contrast, failure to discuss the death fosters avoidance and denial of the finality of the loss (Becker, 1973) . Although many adolescents appear to have the resources needed to cope with the loss of a parent and have no need for interventions from mental health professionals, others do not adapt well to parental loss. This study examines the degree of stress due to parental death reported by adolescents and considers whether certain personal and

PAGE 15

5 family coping resources may moderate the degree of stress resulting from this loss. Scope of the Problem Numerous theories have been proposed explaining the process of adolescent bereavement and adjustment to loss. Models of adolescent bereavement have been developed based on psychodynamic theories of attachment (Bowlby, 1980a; Freud, 1957a, 1957b; Furman, 1974; Parkes & Weiss, 1983), on stage-specific and task-specific processes of grief (Baker, Sedney, & Gross, 1992) , on constructivist theory (Klass, 1988; Rubin, 1985; Rizzuto, 1979; Rosenblatt & Elde, 1990; Silverman, & Worden, 1992) , and on stress theory (Buehler, 1990; Compas, 1987; Compas, Malcarne, & Fondacaro, 1988; Dise-Lewis, 1988; Horowitz et al., 1984; Horowitz, Wilner, & Alvarez, 1979; Lazarus & Folkman, 1984; Patterson, 1988; Patterson & McCubbin, 1987; Spirito, Stark, & Williams, 1988) . Cognitive developmental theorists relate the adolescent's ability to view death as irreversible, natural, inevitable, and personal to their cognitive maturity (Baker et al., 1992; Gordon, 1986; Piaget, 1959; Wass & Stillion, 1988) . Whereas pre-adolescents may view death as irreversible, the cause of death is frequently misconstrued and death is seen as an event of the very distant future. In contrast, an adolescent's thoughts about personal death can become a threat to the emerging identity of the adolescent. According to Erikson's (1959) theory, the

PAGE 16

6 developmental task of adolescence is to separate emotionally from parents and family while establishing an independent identity. During this period of development, the death of a parent may threaten or arrest identity formation, especially in the younger adolescent years where separation has not been completed (Coddington, 1972; Dise-Lewis, 1988; Erikson, 1959; Raphael, 1983; Wass & Stillion, 1988). One model of adolescent bereavement relates to specific stages in which individual tasks of grief are to be accomplished over time (Bowlby, 1980; Furman, 1974; Parkes & Weiss, 1983). The assumption is that if specific tasks are not performed, the person is "stuck" and unable to let go of the deceased and move on with their lives. This linear model is grounded in detachment theory where children are encouraged to cut off ties to the deceased parent, withdraw energy toward the lost love object, and reinvest this energy in new relationships (Freud, 1957b; Furman, 1974). Children are seen as having psychiatric symptoms if they dream about the deceased parent or dwell on their relationship with the deceased. Another model of bereavement, introduced by Baker and colleagues (1992), lists tasks necessary for successful bereavement and places emphasis on cognitive and contextual issues in the grief process. This model provides a shift from the linear stage specific model and suggests that bereavement might be completed over time. This model incorporates the need for family support and identity

PAGE 17

7 development and is an alternative to pathology-based models that identify individuals as being "stuck" in grieving. It promotes a more adaptive view of grief -related behavior and provides direction for short-term therapy goals. Tasks include understanding the fact that someone has died, accepting and bearing the psychological pain of loss, and reorganizing one's identity with a sense of connection to the deceased (Baker et al., 1992). Departure by several researchers from this task oriented model to a cognitive-emotional process model places emphasis on cognitive restructuring of the relationship with the deceased parent (Klass, 1988; Rizzuto, 1979; Rosenblatt & Elde, 1990; Rubin, 1985; Silverman, Nickman, & Worden, 1992) . This approach, grounded in constructivist theory, theorizes that this occurs when the child renegotiates or constructs a new relationship with the deceased parent over time rather than detaching or letting go. As the child moves through developmental stages, the relationship with the deceased parent changes. Thus, bereavement is considered an ongoing process of adjustment (Silverman & Worden, 1992). These theoretical explanations in and of themselves do not explain the differential adjustment of adolescents to the loss experience. A model that emphasizes the role of cognitive mediational processes is the transactional model of stress (Lazarus & Folkman, 1984) with emphasis on cognitive appraisal and coping behaviors (Buehler, 1990;

PAGE 18

8 Compas, 1987; Compas et al., 1988; Dise-Lewis, 1988; Patterson, 1988; Patterson & McCubbin, 1987; Spirito, Stark, & Williams, 1988) . This approach examines the total relationship the adolescent has with the environment to include the environmental context in which the death occurs, the style of coping with stress, and individual and family resources available to the adolescent (Compas, 1987). Adjusting to the death of a parent is a continuous reappraisal process where the adolescent acquires new coping behaviors and resources to meet the demands or changes created by the loss (Dise-Lewis, 1988; Patterson, 1988). The relationship between personal resources and coping behaviors is of interest to researchers in determining the type of coping strategies adolescents use to manage stress. The adolescent's repertoire of personal resources such as inherent traits, abilities, and means utilized to meet demands are knowledge, skills, personality traits, emotional and physical health, self-esteem, and self -concept (Patterson & McCubbin, 1987; Jackson & Bosma, 1990) . Three types of coping behaviors have been cited in the literature. First, direct actions to reduce or eliminate demands, referred to as problem-focused coping, are one type. A second type consists of behaviors directed at redefining demands to make them more manageable, termed appraisal-focused coping behaviors. A third type, emotionfocused coping, refers to behaviors directed at managing the

PAGE 19

9 tension that is felt from experiencing demands (Patterson & McCubbin, 1987) . Researchers have reported that adolescents who use more approach-oriented, problem-focused strategies to cope with interpersonal stressors have fewer emotional and behavioral problems, whereas those who use more avoidanceoriented, emotion-focused strategies have more problems and are more likely to adjust poorly when confronted by subsequent life crises and stressors (Compas et al., 1988; Ebata & Moos, 1991; Harris, 1991). Different phases of adolescent development are thought to influence the selection and use of particular coping strategies. Younger teens seem to have fewer resources in coping with parental death. They may report overpowering memories of the deceased and preoccupation with the loss. Frequently, older teens use denial and avoidance to escape from dealing with the death. Stress related symptomatology is prominent with both groups, and grief resolution is compromised with these coping methods (Harris, 1991) . Investigators report that adolescents identify peers as being the most helpful to them after parental death and are more likely to talk with siblings and friends about their feelings than talk with parents (Cragg & Herman, 1990; Gray, 1989; Silverman & Worden, 1992). Others have found that the need for adolescents to fit into a peer group may inhibit expression of grief for fear of being singled out as different from peers (Elkind, 1968; Harris, 1991).

PAGE 20

10 The most frequently cited family resources in the research literature reported to facilitate stress management following parental death are cohesion, parent-adolescent communication, and adaptability (Barnes & Olson, 1985; Buehler, 1990; Cragg & Herman, 1990; Gray, 1987; Olson, Portner, & Bell, 1982; Silverman & Worden, 1992). High family cohesion and social support are reported by adolescents to be protective against depression (Balk, 1991; Gray, 1987; Walker, 1985). There is, however, a systematic tendency for adolescents to describe their families as demonstrating less cohesion and less adaptability than their parents report on these values (Noller et al., 1992). Clinical literature on childhood bereavement stresses that open communication between parents and children greatly facilitates the mourning process (Bowlby, 1963; Cain & Fast, 1966; Furman, R.A. , 1964; Furman, E. , 1974, 1983; Kliman, 1979; Miller, 1971; Nagera, 1970) and protects youth from the onset of psychopathology in later life (Cheifetz et al., 1989) . The ability of each family member to express thoughts and feelings about the death and not be inhibited by one another's emotions is thought to relieve stress and is considered to be "open communication" (Brown, 1988) . Adult studies of childhood parental death report that open communication with the surviving parent around issues related to the death of a parent is an important factor in long-term adjustment and protection against depression. In adolescent studies, researchers find that communication is

PAGE 21

11 often disrupted in bereaved families (Silverman & Silverman, 1979). The surviving parent and children often avoid discussing the death to spare each other pain. Parents are generally overwhelmed by their own grief and are unavailable to help their own children grieve effectively. Adolescents are not able to accept parental support because they fear it will compromise their needs for autonomy (Adams-Greenly & Moynihan, 1983; Gray, 1987; Harris, 1991; Raphael, 1983). Family systems theorists report that silence about the death encourages a closed family system and family communication patterns become distorted and rigid (McGoldrick & Walsh, 1991) . The death of a parent creates extreme stress within the family. Anxiety is high and attempts are made to protect oneself and other members from the emotional pain of loss. Instead of communicating feelings to other members of the family for emotional support, members fear upsetting others. There is a clear message that members of the family are not allowed to talk about the death or express feelings (Kerr & Bowen, 1988) . Crying is discouraged and emphasis is on "getting on with life" and detaching from the deceased. This closed system is unable to integrate the loss and incorporate grieving into its identity while grief is suppressed and denied. Reorganization for the family becomes rigid and dysfunctional and roles are rigidly maintained (Crosby & Jose, 1983; Davies, Spinetta, Martinson, McClowry, & Kulenkamp, 1986) . Members are chosen to fill the role of

PAGE 22

12 the deceased, and often a child is chosen to take on the role of the deceased adult. There is little research in the area of adolescent adjustment to parental loss and how the impact of the family unit affects reorganization and adaptability after the death of a parent. Gehring and Feldman (1988) define adaptability as the ability for the family system to change its power structure of roles and rules in response to situational and developmental stress. Adjustment is found to be greatly determined by the ability of the surviving parent to maintain consistency and stability in the adolescent's environment (Siegel et al., 1990). The functioning family is defined as a flexible unit of individual members who reorganize after the death where no member is expected to fill the role of the deceased parent. Less functional families are described as being resistant to reorganization and they frequently place a child in the role of the deceased parent (Crosby & Jose, 1983; Davies et al., 1986). Theoretical Framework At the heart of an exploration into an adolescent's response to parental loss is the conceptualization of the stress experience. Life event theorists report that it is the quality of the life event and circumstances surrounding it that determine the degree of stress incurred by the event (Pearlin & Schooler, 1978). Horowitz et al. (1993) describe the death of a loved one as a life event that evokes stress response syndromes. They describe stress response syndromes

PAGE 23

13 as recurrent intrusive images, ideas, and feelings related to a previous traumatic experience. Oscillations between episodes of intrusion and denial are seen as evidence of incomplete personal schema for understanding death and reacting to this event. Horowitz' explanation of intrusion is based on Freud's (1957b) ego-psychological explanation of compulsive repetition of warded-off ideas and associated emotions as an attempt of the ego to master and assimilate traumatic experiences. The repetitive motive is described by Horowitz as a "completion tendency" or a cognitive motive used to reduce the discrepancy between new information elicited by the life event and pre-existing inner models of meaning or schemata. This discrepancy is resolved through the process of assimilation and accommodation defined by Piaget (1959). As long as the memories associated with the life event are held in the active memory processes, it is likely that the information can be processed and assimilated into pre-existing schemata for completion. However, traumatic memories might activate dreaded states of mind that are too emotional, too painful, too out of control, and too hopeless. The memories might then become avoided. Horowitz et al. (1979) developed an instrument, the Impact of Event Scale, which measures intrusion and avoidance in response to a life event. Intrusion is represented by the involuntary awareness of ideas, memories, and emotions associated with the event, and avoidance is represented by the conscious attempt to divert >

PAGE 24

14 attention away from cognitions and feelings related to the event. The instrument is designed to measure subjective experience in response to a life event over time and has been used in recent research to measure the impact of the death of a parent on adolescents (Harris, 1991). An application of a transactional model of stress to bereaved adolescents is a dynamic conceptualization that emphasizes the relationship between the adolescent and the environment. According to this theory, stress occurs when the individual perceives demands of a situation as taxing or exceeding available coping resources to meet those demands, especially when the system's well-being is perceived as being at stake. It is neither a characteristic of the environment alone nor of the person alone. Rather it is a relational, process-oriented task between the individual and the environment for cognitive-affective appraisal and coping activities. Cognitive appraisal means that the individual has the knowledge or beliefs about how things work and the ability to evaluate the personal significance of encounters with the environment. Appraisal components serve different functions and are influenced by individual development and sociocultural variables. Primary appraisal evaluates stressful environmental conditions to determine the nature of harm/loss, threat, or challenge. Secondary appraisal evaluates which coping options are available and can be used effectively. Person-environment interactions are constantly

PAGE 25

15 evaluated and adjustments are made by coping through this reappraisal process. Coping is viewed as a problem-focused or emotionfocused planned action that regulates emotional distress and shapes subsequent emotion. Problem-focused coping is an attempt to act on the stressor and change the actual personenvironment relationship by directly acting on the environment or on the self. Emotion-focused coping alters the appraisal component of the mind through avoiding the stressor or by cognitively reframing the meaning of the relationship through denial or distancing, thus making the distressing emotion moot. If the meaning of an event is changed, the emotion will also be changed (Lazarus, 1991b) . Younger adolescents are reported to use more emotion-based types of coping such as avoidance, denial, and self -blame, whereas older adolescents are reported to utilize direct action as coping methods (Stern et al., 1992). In addition to problem and emotion-focused coping skills, the presence of a supportive family environment provides resources for successful adjustment to loss in adolescents (Compas, 1987). The integration of individual coping theory with family stress theory provides constructs to examine individual and family resources for adjustment to stress. The most frequently assessed dimensions of family functioning in family research include family cohesion, adaptability, organization, parent-adolescent communication, and conflict resolution skills (Adams-Greenly & Moynihan,

PAGE 26

16 1983; Buehler, 1990; Cragg & Herman, 1990; Gray, 1989; McGoldrick, 1991; Moody & Moody, 1991; Noller et al., 1992; Siegel et al., 1990; Silverman & Silverman, 1979; Stern et al., 1992). The degree to which family functioning and personal coping skills are instrumental in determining adjustment to the stress of parental death is unknown. The assessment of personal and family resources has the potential of being a helpful diagnostic tool in assessing those adolescents who may be particularly vulnerable to adverse effects of parental loss. Need for the Study Little is known about the resources and coping skills that adolescents use to manage the stress created through parental death (Parkes & Weiss, 1983). Research on adolescent bereavement has typically been presented using standardized measures for depression and psychopathology . These measures fail to depict grief-related symptoms as a normal process of bereavement. Normal bereavement needs to be characterized, including risk factors, and special attention given to developmental stages of adolescent development in relation to grief reactions (Harris, 1991; Fleming & Adolph, 1986) Bereaved adolescents may be at risk because they do not have independent access to treatment as do surviving spouses. Surviving parents may be overwhelmed by their own

PAGE 27

17 grief and may not recognize the need for intervention with their children (Harris, 1991) . Many of the studies have relied on parent reports of adolescent response to death which precludes measurement of those experiences unknown by the parents (Dise-Lewis, 1988; Elizur & Kaffman, 1982; Silverman & Worden, 1992; Vida & Grizenko, 1989), and a number of studies have shown that parental measures often correlate to a lesser extent with adolescent self-report measures (Swearingen & Cohen, 1985). Adolescent coping strategies have been measured with predetermined lists rather than open-ended measures and response to stress due to parental death has not been charted over a long period of time (Compas, 1987) . Age appropriate coping measurements have not been used to reflect changes in cognitive development and response capabilities with parental death. Thus, there is a need to examine a wide range of adolescents differing in their patterns of adaptation to parental loss. Understanding the impact of coping resources of bereaved adolescents could provide important information about the intervention needs of adolescents. Purpose of the Study The purpose of this study is twofold. First, the degree of stress reported by bereaved adolescents will be examined. Second, using a transactional stress model, the mediating influences of the adolescent's personal coping strategies and family life on their stress response will be

PAGE 28

18 explored. Three aspects of adolescent family life will be assessed: (a) family adaptability, (b) family cohesion, and (c) parent-adolescent communication. Research Questions In this study, the relationships between stress, the quality of family life, and personal coping strategies will be described as they pertain to adolescents who have experienced the death of a parent. Specifically, adolescents will be described in terms of the degree of stress experienced, the types of coping strategies utilized, the level of family cohesion, the level of family adaptability, and the degree of positive parent-adolescent communication. The specific research questions are as follows. 1. After controlling for age and length of time since the death, how do adolescents' use of certain coping strategies relate to their reported degree of stress? 2. After controlling for age and length of time since the death, how do adolescents' perceptions of their extent of positive parent-adolescent communication relate to their reported degree of stress? 3. After controlling for age and length of time since the death, how do adolescents' perceptions of the level of family cohesion relate to their reported degree of stress? 4. After controlling for age and length of time since the death, how do adolescents' perceptions of their

PAGE 29

19 degree of family adaptability relate to their reported degree of stress? Significance of the Study Because certain person-based and family-based strategies appear to be useful to adolescents coping with bereavement, focusing on the role they play in mediating adolescent stress would seem like a relevant path to take. Specific types of family coping styles would be relevant to look at as well as personal coping abilities. Definitions Family adaptability is defined as the extent to which the family system is able to change its power structure, role relationships, and relationship rules in response to situational and developmental stress (Olson et al., 1992). Adolescence in this study is defined as the period of development between 12 and 18 years of age. Adolescent adiustment is determined by the level of stress evident as measured by the Impact of Event Scale after the death of a parent. Avoidance is a psychological reaction to stress and is part of the stress response characterized by ideational constriction, denial of the meanings and consequences of the event, blunted sensation, behavioral inhibition or counterphobic activity, and awareness of emotional numbness (Horowitz et al., 1979). Cognitive-affective appraisal is an appraisal based on the knowledge or belief about how things work and the

PAGE 30

20 ability to evaluate the personal significance of encounters with the environment (Lazarus, 1991b). Family cohesion is defined as the emotional bonding that family members have toward one another. Within the Circumplex Model, cohesion is measured by emotional bonding, boundaries, coalitions, time, space, friends, decisionmaking, interests and recreation. (Olson et al., 1992). Family communication is defined as the extent to which the adolescent expresses feelings directly and shares factual and emotional information with their parent as measured by the Adolescent-Parent Communication Scale (Barnes & Olson, 1985) . Coping is defined as any activity that regulates emotional distress and shapes subsequent emotion. It is a planned action that is either problem-focused or emotionfocused (Lazarus, I99la) . Emotion -focused coping is defined as any activity designed to alter the appraisal component of the mind through avoidance or by changing the meaning of the relationship through denial or distancing (Lazarus, I99la) . Intrusion is a psychological reaction to stress and is part of the stress response characterized by unbidden thoughts and images, troubled dreams, strong pangs or waves of feelings, and repetitive behavior (Horowitz et al., 1979) . Parent , for the purposes of this study, is defined as any person who is the primary caretaker and legal guardian

PAGE 31

21 of the adolescent. This person is not necessarily the biological parent. Problem-focused coping is an attempt to change the actual person-environment relationship by directly acting on the environment or on the self (Lazarus, 1991a) . Psychological stress is a result of a relationship where environmental demands tax or exceed resources of the individual and negative emotions are present based on harm/ loss, threat, and challenge (Lazarus, 1991a) . Stress response is the psychological response to stressful life events by intrusion and avoidance (Horowitz et al. , 1979) . Transactional theory of stress defines stress as a particular type of relationship between the person and the environment, in which the individual perceives demands of a situation as taxing or exceeding available coping resources to meet those demands, especially when the system's wellbeing is perceived as being at stake. Organization of the Study This proposal consists of three chapters. Chapter 2 consists of a review of the related literature. Chapter 3 presents the methodology, which includes a description of the research design and variables, the population, sampling procedures, instrumentation, data collection and research questions. Chapter 4 presents the results of the statistical analyses of the data. Chapter 5 includes a discussion of the results, implications for therapy.

PAGE 32

22 limitations of the study, and suggestions for further research.

PAGE 33

CHAPTER 2 LITERATURE REVIEW Introduction In this chapter, a review and synthesis of the literature relevant to adolescent stress response to parental death will be presented including theoretical perspectives on stress response and personal and family coping resources. Impact of Pa rental Death on the Adolescent The majority of literature available explaining adolescent stress response to parental death utilizes a developmental lens. Differences in adolescent stress responses are posited to be shaped by the child's emotional and cognitive processing capacities. Piaget (1959) , for example, suggested that adolescents reach a stage of cognitive development or formal operational thought when they are able, for the first time, to grasp abstract principles and ideas. This cognitive capacity is thought to introduce a new concept of death where death is understood as an irreversible, natural, inevitable, and personal phenomenon (Baker et al., 1992; Gordon, 1986; Wass & Stillion, 1988). The adolescent becomes capable of making a subjective appraisal of the meaning of death, the demands created by the loss, and the personal resources available to

PAGE 34

24 cope with the loss. According to stress theory, death becomes a stressor when the perception of the demands created by the death are outweighed by the perception of capabilities available to meet the demands. The impact of parental death is viewed by the majority of researchers as the most profound event shaping the subsequent development of the adolescent. Erikson (1959) proposed that the primary task of adolescent development is to develop an identity and role in society separate from parents and family. Raphael (1983) suggested that the loss of a father or mother during this stage is considered disruptive to normal developmental tasks, instills a fear of lack of control over life events, and creates a threat to the adolescent's mortality. New demands are placed on the adolescent while adjusting to the loss, and the ability to manage the stress created by the death is contingent upon available personal and family resources. Numerous studies have investigated the effects of parental death on various indices of change in the lives of adolescents. Family relationships change with parental death as well as perceptions of personal and family identities. Increased family responsibilities, social isolation from friends, lack of communication with surviving parents, economic insecurity, and identity crisis due to premature separation from parents all contribute to the stress created through parental loss.

PAGE 35

25 Cragg and Berman (1990) conducted a study of ten adolescents, six months to two years after parental death. They found role reversals in the families as adolescents assumed more household responsibilities and assumed the role of protecting parents who were preoccupied with grief. Some were pressured by family members to take on the role of the deceased parent to help the family avoid the pain of mourning. Adolescents reported social isolation from friends and felt that only those who had lost a parent could understand their feelings. There was reluctance to talk with the surviving parent about their feelings, and a decrease in socioeconomic status was found with one family. Social isolation from friends has also been described by Gapes (Unpublished manuscript, 1982) in a study of bereaved adolescents who felt that their emotions were unacceptable to members of their social environment and especially to their peers. The adolescents reported that they did not know what was normal in a grief reaction and were hesitant to show their natural responses for fear of being noticed and considered different or abnormal. The typical response of adolescents was one of "escape" by suppressing emotions compared to their surviving mothers who were described as "preoccupied" with mourning. Adolescents felt deficient in expressing emotions, uncertain about the meaning of death, and unsupported by others. Gray (1989) studied 50 parentally bereaved adolescents between 6 months and 5 years after the death and

PAGE 36

26 found a lack of communication between adolescents and surviving parents. Adolescents reported better understanding by peers than family members and felt more comfortable expressing their feelings outside the family. Peers were rated as the "most helpful" by 80% of the group. Most reported that they relied on only a single friend or a small group of friends for help. Adolescents valued being treated in usual ways and not as a "different" person because they had lost a parent. Some parents were overwhelmed by their own grief and could not help their children. Baker et al., (1992) suggested that during the early stage of grief, there is a desire to contain the pain of mourning and limit change as much as possible. They imply that there is an intense self-protectiveness of the bereaved individual or family against the full emotional significance of the loss. It is suggested that those who have the least amount of change in their lives adapt better to the stress of parental loss. Normal Stress Re sponse to Parental Death Stress response to the death of a parent has been described by researchers in terms of behavioral, emotional, and physiological reactions. The initial response to parental death has been described as one of shock, numbness, and disbelief (Raphael, 1982; Silverman & Worden, 1992). Gapes (Unpublished manuscript, 1982) defined the adolescent's typical bereavement response as one of "escape"

PAGE 37

27 and identified responses of guilt, anger, depression, disorientation, and insecurity. Research studies of normal, non-clinical populations have been identified to represent the normal stress response to parental death. Cragg and Herman (1990) identified several indicators of stress, with a non-clinical study of adolescents whose parents died within the previous 6 months to 2 years, such as decreased family communication, increased household responsibilities, increased nightmares, feelings of being different from their friends, thinking that friends could not understand their feelings, and reluctance to talk to parents about the death. Silverman and Worden (1992) observed normative behaviors 4 months after parental death in a sample of nonclinical adolescents as follows: crying, insomnia, learning difficulties, health problems with somatization of feelings, headaches, difficulty concentrating in school, external locus of control, and lower levels of self-esteem. Gray (1987) reported higher depression scores with bereaved adolescents compared to nonbereaved adolescents and decreased academic performance for younger adolescents indicating that cognitive functions may regress temporarily during bereavement. In a review of childhood bereavement literature, Vida and Grizenko (1989) found no specific syndrome associated with bereavement, but found an increase in features such as sadness, crying, irritability, minor depressive syndrome.

PAGE 38

28 temper tantrums, withdrawal, insomnia, appetite problems, enuresis, and learning and academic problems. The authors suggested that bereavement studies have failed to establish the boundaries between childhood bereavement, major depression, and adjustment disorder. Pathological Stress Response to Parental Death Raphael (1983) suggested that adolescent grief may become pathological due to ego vulnerability. Fears of regression or loss of emotional control may prevent expression of normal feelings of loss and create ambivalent relationships with family and friends. Consequently, adolescents are thought to act out solutions to inner stress often in the manner in which grief reactions are discharged while grief remains unresolved. In case studies of bereaved adolescents, Raphael (1983) found that patterns of suppressed or inhibited grief response, distorted grief, and chronic grief create behavior problems and maladjustment. These patterns were found to create some of the following outcome behaviors: school delinquency, increased alcohol and drug intake, phobias, suicide ideation, depression, antisocial behaviors, academic failure, stealing, fighting, reckless driving, unwanted pregnancy, other risk-taking behaviors, and hostility toward the surviving parent. Chiefetz et al., (1989) conducted a clinical study of adolescents, from 11 to 17 years of age, whose parent died 2 years prior to the study. Subjects were referred to an outpatient psychiatric service with symptoms of mild

PAGE 39

29 depressive affect with dysthymic disorder. Their depressed mood consisted of loss of interest or pleasure in almost all activities for a period of one year. Elizur and Kaffman (1983) also found evidence of dysthymic disorder in a study of parentally bereaved adolescents of 12 years and older. In the absence of overt symptomatology, another type of pathological behavior has been observed in older adolescents of the same sex as the surviving parent in the form of supportive and extremely protective behaviors of the surviving parent (Elizur & Kaffman, 1983). The behavior was observed more often during conjoint treatment sessions with the adolescent and surviving parent. Bowlby (1980) described similar behaviors as "compulsive care giving" of the surviving parent where the adolescent exhibits exemplary behavior as a defensive facade in the absence of personal grieving. This type of behavior is associated with prolonged mourning. Several factors that may contribute to pathological stress response to parental death that are not controlled for in studies of bereavement are pre-existing emotional environmental influences in the family, the circumstances of the death, reactions of the surviving parent, and previous life experiences of the adolescent (Cheifetz et al., 1989). Factors such as parental depression and alcoholism are also likely to be factors associated with a more difficult bereavement.

PAGE 40

Shapiro (1994) suggests that violent deaths or suicide of parents as well as previous traumatic experiences may contribute to pathological stress response in adolescents. Trauma reactions may accompany, interfere with, or mask grief reactions. Eth and Pynoos (1985) suggested that in the case of traumatic grief reactions, the process of dealing with the traumatic experience takes priority over dealing with the loss. In a study of a violent death where the child has witnessed the physical mutilation of a parent, the fear aroused by the violent image may interfere with exploration of reminiscences that over the normal course of grief, allow identification with the dead parent (Eth & Pynoos, 1985). Lukas and Seiden (1987) found that when the murderer is a stranger, personal safety and fear interfere with the process of grief. When the murderer is a family member, conflicts of loyalty as well as rage occur. With suicide, the child suffers intense guilt in addition to resentment if the child felt responsible for keeping the parent alive (Lukas & Seiden, 1987) . From a review of the literature, there appear to be distinct features of parentally bereaved adolescents. Studies suggest that there is a threat to adolescent identity as motherless or fatherless adolescents adjust to who they are without one of their parents. The possibility of role reversals exist where adolescents are pressured to take on responsibilities of the deceased in order to

PAGE 41

31 maintain some degree of normalcy in the family. Surviving parents may be preoccupied with grief and unable to take care of their families. Communication problems are reported, and there are conflicting reports in the literature about adolescents' willingness to share feelings with family members or peers about their response to the death. There is evidence that those adolescents who are able to process their grief by talking about feelings and the events surrounding the death are able to grieve successfully. Those who are unable to process their grief and feel socially isolated from family and peers seem to have difficulty coping with the loss. Stress response in adolescents has been distinguished in the literature through clinical and nonclinical populations. It appears that there is a typical grief reaction to parental death and that it is expressed by initial shock, numbness, and disbelief. Nonclinical population studies are thought to represent normal responses to grief and include behavioral, emotional, and physiological reactions to parental death. In contrast, stress response is usually considered pathological in cases of complicated grief where expression of grief is suppressed, inhibited, distorted, or when grief is expressed in chronic behavior patterns that create problems for the adolescent. In summary, symptoms of bereaved adolescents are specified as normal or pathological depending on the

PAGE 42

32 intensity of stress, the duration of maladjustment, and the extent of adaptability to change. There is no clearly defined bereavement syndrome of adolescents described in the literature that allows the researcher or clinician to clearly define the boundaries between bereavement stress response, depression, and adjustment disorder. Theoretical Models of Childhood Bereavement There are several models of childhood bereavement in the literature that describe cognitions and behaviors relative to normal and pathological reactions to the death of a parent. The models are based on detachment theory, stage-specific and task-specific processes of grief, and constuctivist theory. Theorists traditionally encouraged the bereaved to disengage from the deceased and let go of the past (Dietrich & Shabad, 1989; Furman, 1974, 1983; Volkan, 1981; Miller, 1971). This approach was grounded in Freud's (1957b) detachment theory which suggests that in order to successfully resolve grief, a person must gradually withdraw mental energy extended toward the lost love object and reinvest the energy in new relationships. Failure to accomplish this detachment was thought to result in symptoms of psychological problems. This theory was conceptualized from clinical case studies with troubled subjects who were more likely to attach inappropriate emphasis on relationships with the deceased (Silverman et al., 1992). Bereavement was seen as a psychological process that was to

PAGE 43

be completed or resolved in order to adjust. Other models evolved to describe the process further. Stage specific models were developed where the grief process was explained by a description of emotions from numbness, to yearning, to despair, and reorganization (Bowlby, 1980a; Parkes & Weiss, 1983) . Problems were thought to occur when individuals became "stuck" in a stage and were unable to progress to the next stage until they had accomplished tasks of the immediate stage. Bereavement was considered an orderly, linear process that could be successfully completed in sequential stages. There are limitations in applications to treatment with this model because it merely gives a picture of the process at different points in time and does not specify what is required for the individual to move through the stages. Task specific models have been cited in the literature and used by Furman (1974) and Shuchter (1986). A time-specific task model was presented by Baker et al. (1992), which outlined specific psychological tasks to be performed over time where internal attachment to the deceased was considered a healthy recovery, not a sign of pathology. This three-phased model first requires that the child have the cognitive ability to understand death and to feel safe in a secure environment. The middle phase requires the child to accept the pain of loss and reevaluate the relationship to the lost object. During the last phase, a reorganization of the child's sense of identity and of

PAGE 44

34 significant relationships is required, which allows for a new sense of identity while acknowledging the loss and maintaining some sense of identification with the deceased. This model gives emphasis to developmental stages of the child and to the importance of the role of the family in helping the child process the grief of parental loss. The drawback in this approach is the tendency to propose overly ambitious treatment goals for bereavement; and although the concept of regression during developmental transitions is part of the theory, the implication is that the process may be completed over time. Also, case studies presented with clinical subjects were of troubled children. Silverman et al. (1992) conducted a child bereavement study of non-clinical children aged 6 to 17 years who had experienced a parental death. Bereavement was considered a cognitive-emotional process whereby the child continued to remain in a relationship with the deceased. This concept was a shift from the former view of bereavement as a psychological state that ends or from which one recovers. Emphasis was on the negotiation and renegotiation of the meaning of loss over time, rather than letting go of the lost love object. This observation is supported in the literature by Rizzuto (1979) who observed that the process of constructing inner representations of the deceased changes with development and maturation. Rosenblatt and Elde (1990) also found that grief work included maintaining connections with memories of the deceased while integrating these

PAGE 45

35 memories into the present and into relationships with others. Family members were found to be helpful to each other through shared representations of the deceased. These studies are grounded in constructivist theory where individuals are regarded as constructing their own reality (Gergen, 1985) . The bereaved child is thought to construct an inner representation of the deceased parent through memories, feelings, and actions such that the child remains in a relationship with the deceased. This relationship changes as the child matures and accommodates to the loss, developing a new perspective of the meaning of the relationship with the deceased. Stress Theories Stress theorists initially referred to stress as an analogy of Hook's 17th century principles of engineering. Stress was considered the ratio of the force of the external stimulant that put strain on an object to the deformity of the object. Emphasis was on the external environment rather than what was happening in the mind. Cannon (1939) and Selye (1956) viewed stress as the physiological disturbance in response to the strain. Selye later coined the term "general adaptation syndrome" which stated that organismic stress was the same regardless of the type of stressor. Focus was on an extreme response to stimuli with the exact nature of the stressor seen as irrelevant to the response. Life events theorists looked at specific life experiences, and inventories were developed to measure "life

PAGE 46

36 change units" (Coddington, 1972) to provide an index of the amount of change resulting from the degree of stressfulness associated with experiencing specific events (Johnson, 1986) . Not only were specific experiences held accountable for the degree of stress but the frequency and similarity of stressful events were found to increase vulnerability to having subsequent stress reactions in response to stressors of the same type (Horowitz et al., 1979). A more global theory of stress was developed by Lazarus (1966) , which suggested that the stressfulness of events not only depends on occurrence of events but on the child's appraisal of the meaning of the events. This cognitive appraisal approach proposed that specific kinds of information determined if a particular stimulus was considered a stressor. Lazarus & Folkman (1984) described appraisal and coping as mediators of stress emphasizing individual differences in reaction to common stressors. This theory further developed into a cognitive-motivational-relational theory of emotion known as a transactional model of stress which provided a broader conceptualization of the interpretation of life events as potential stressors. Life Events Stress Model One of the initial inventories which measured life events as units of change or stress for children and adolescents was Coddington's (1972) Life Events Record. The amount of social readjustment required by specific events determined the degree of stress incurred. In a study of

PAGE 47

37 3,526 children ranging from preschoolers to senior high school students, Coddington found a strong relationship between the amount of life change experienced and the age of the child, with older children and adolescents experiencing higher levels of life change than younger children. The death of a parent was more than 20 units higher than the next highest stressor for the participants. Dise-Lewis (1988) developed a comprehensive measure of life stress for children based on life events and coping strategies generated by child subjects. The amount of life changes experienced by 681 normal children, the perceived stressfulness of these changes, and the types of coping strategies used by the children to manage stress were incorporated into the Life Events and Coping Inventory. Parental death was rated as the most stressful event in the lives of the subjects. The researcher found that children begin to set a pattern of stress responses by early adolescence and that stress scores correlate well with established measures of anxiety, depression, psychosomatic symptoms, and behavior problems. Horowitz et al. (1979) studied subjective stress in response to death and constructed a scale, the Impact of Events Scale, to measure stress response as it is expressed through intrusion and avoidance. Intrusion includes involuntary recurrent intrusive images, ideas, and feelings related to the death experience. Avoidance was described as the voluntary, conscious attempt to divert attention from

PAGE 48

38 the images, ideas, and feelings related to the experience. He found oscillations between episodes of intrusion and avoidance to be common after the death and attributed this to incomplete personal schema for understanding death and personal reactions to the event. Horowitz, Field, & Classen (1993) explained the motive to repeat the experience as a completion tendency which is based on Freud's (1957) ego-psychological explanation of the compulsive repetition as an attempt of the ego to master and assimilate traumatic experiences by repeated review of the event and its implications. They further explained intrusion as a cognitive attempt to reduce the discrepancy between new information or implications elicited by the event in active memory to pre-existing models of meaning or schemata. The idea was that the new information would eventually be processed to the point where there was no discrepancy and the new information would be assimilated into pre-existing schemata. Traumatic memories and emotions associated with the death were thought to prolong or prohibit assimilation. The ego may not be able to accept the threat of traumatic events and avoidance may be used to defend the person from dreaded states of mind that are too emotional, too painful, too out of control, and too hopeless (Horowitz et al., 1993). Harris (1991) administered the Impact of Events Scale to 11 healthy adolescents between the ages of 13 and 18 years of age at 1 and 6 months after parental death. All

PAGE 49

39 subjects reported moderate to high levels of intrusive and avoidant stress-related symptomatology at the initial assessment with the Impact of Events Scale. Symptoms decreased over the year, but more than half of the subjects continued to experience moderate or high levels of distress 1 year later. Horowitz et al. (1986) proposed a treatment of stress response syndromes for normal and pathological phases of poststress response. A pattern of phases that individuals tend to progress through in normal responses to serious life events was identified by Horowitz and pathological phasic response to extreme stress was thought to be an intensification of the normal phasic response. He suggests that through a collaborative relationship with the therapist, the patient recounts circumstances of and surrounding the traumatic event. This process involves a reappraisal of the event, the meanings associated with it, and preexisting features of the personality of the patient. The goal is to move the patient from an overcontrolled orientation or undermodulated intrusive reaction to a manageable exploration of the meaning the traumatic event has for the patient. Transactional Stress Model According to this theory of stress, psychological stress is a result of a relationship where environmental demands tax or exceed resources of the individual and negative emotions are present based on harm-loss, threat,

PAGE 50

40 and challenge. Stress is considered a relational, processoriented task between the individual and the environment with a goal for cognitive-affective appraisal and coping activities. There are different appraisal processes which serve different functions. Primary appraisal evaluates stressful environmental conditions to detemnine the nature of stress. Secondary appraisal evaluates which coping options are available and which can be used effectively. Person-environment interactions are constantly evaluated and adjustments are made by coping through this reappraisal process (Lazarus, 1991b) . Coping is an activity that regulates emotional distress and shapes subsequent emotion. It is a planned action that is considered either problem-focused or emotionfocused coping. Problem-focused coping is an attempt to change the actual person-environment relationship by directly acting on the environment or on the self. Emotionfocused coping alters the appraisal component of the mind through avoidance or by changing the meaning of the relationship through denial or distancing, making the distressing emotion moot. If the meaning of an event is changed, the emotion will also be changed (Lazarus, 1991b). Compas et al. (1988) applied Lazarus' theory of coping to a study of 130 children from 10 to 14 years of age. They were given an open-ended instrument to assess coping with self -identified recent stressful events. Illness of a friend or family member was frequently listed

PAGE 51

41 as a social stressor. The authors found that the number of emotion-focused strategies used with social stressors increased with the age of the child while the use of problem-focused coping was relatively consistent with all age groups. In developing the instrument, Adolescent Coping Orientation for Problem Experiences (A-COPE) , Patterson and McCubbin (1987) conducted a 3-year longitudinal study designed to examine the influence of family variables on adolescent substance use. Of the subjects, 34% were between the ages of 14 and 18. Gender differences in types of coping were found where females had more frequent use of coping behaviors direciied at developing social support than males. It was found that females report more frequent use of a broader range of coping patterns than males. A-COPE was found to self-educate adolescents about their coping styles and options and to be helpful in teaching adolescents about managing stress. Adolescent Coping Models Folkman and Lazarus (1980) viewed coping as the effort to manage specific events or stressors that have been appraised as demanding and as taxing the individual resources. They designed several models to classify coping strategies in adolescents and to explain the relationship between coping and adjustment. The approach/avoidance coping model conceptualizes coping in terms of active versus passive strategies in which

PAGE 52

42 active cognitive attempts are used to change the ways of thinking about a problem and behavioral attempts are used to directly resolve the problem. Avoidant strategies include cognitive attempts to deny or minimize threat and behavioral attempts to get away from the problem situation (Billings & Moos, 1981; Lazarus & Folkman, 1984; Roth & Cohen, 1986). Another model which classifies coping responses according to their function is the problem/emotion-focused coping model (Folkman & Lazarus, 1980; Menaghan, 1983). Coping strategies are problem-focused attempts to modify the stressor or emotion-focused attempts to regulate emotional states that accompany stress. Compas et al. (1988) found that adolescents who used more problem-focused strategies to cope with interpersonal stressors, like talking things over with others, had fewer emotional and behavioral problems, and those who used more emotion-focused strategies like avoidance or yelling had more emotional and behavioral problems as measured by the Child Behavior Checklist (CBCL) and tne Youth Self Report version of the CBCL. Patterson and McCubbin (1987) derived a scale of coping responses reported by adolescents which identified 12 direct action and indirect coping response patterns of behavior used when adolescents were faced with difficulties or felt tense. In studies on coping strategies and adolescent substance use, indirect coping through reliance on peer support and ventilating feelings was related to greater substance use, whereas direct coping by seeking

PAGE 53

43 greater family support and seeking adult support was related to less substance use (Patterson & McCubbin, 1987; Wills, 1986; Dise-Lewis, 1988; Spirito et al., 1988). In sununary, it appears that the effectiveness of adolescent coping varies according to the stage of development, appraisal of the stressor, and coping style. Person-Based Cooing Resources Development Adolescents are often assumed to be at high risk for psychological stress by developmental stress theorists because they are confronted with many life stressors for the first time without a repertoire of coping responses from which to draw (Konopka, 1980). Spivack and Shure (1982) found that the ability of adolescents to generate a variety of alternative solutions to interpersonal problems is related to behavioral adjustment. Alternatively, youth with emotional and behavioral problems were found to generate fewer alternative solutions than control subjects. Lazarus and Launier (1978) found that dysfunctional coping is primarily associated with withdrawal, although withdrawal has also been found to be a meaningful way of coping with situations of extreme stress or where direct action is inhibited by external barriers. Age of development, personality, and perception of the stressor have been found to be determinants of the particular styles of coping (Tyszkowa, 1991) . Older adolescents have been found to utilize more active problem-solving coping

PAGE 54

44 behaviors than younger adolescents who use more emotionbased types of coping (Compas et al., 1988; Patterson & McCubbin, 1991). Appraisal Holroyd and Lazarus (1982) define appraisal as the evaluative process the individual uses to recognize threat or harm and then determine the resources and options available for managing potential or actual harm. Maddi (1981) differentiated between two forms of appraisal that used avoidance and transformational coping. Avoidance coping implies pessimistic cognitive appraisal and evasive action, whereas transformational coping implies more optimistic appraisal and deliberate actions to alter or reduce stressful events. In adult studies of the relationship between coping and cognitive appraisal of stressful situations (Folkman & Lazarus, 1980), it was found that problem-focused coping was used in situations that were appraised as changeable and more emotion-focused coping was used in situations that were appraised as unchangeable realities that must be accepted. Forsythe and Compas (1987) further discovered that psychological symptoms were related to the degree of fit between appraisals of control and the relative amount of problem-focused and emotion-focused coping used. Psychological symptoms occurred when subjects used more emotion-focused coping with events perceived as controllable and more problem-focused coping with events perceived as uncontrollable.

PAGE 55

45 In a study with adolescents in grades 6 through 8, Compas et al., (1988) found that adolescents match their coping to fit their appraisal of control over stressful events. Coping strategies were observed with academic stressors such as receiving a poor grade and with interpersonal stressors such as illness of a friend or family member. Females reported using more emotion-focused strategies than males with interpersonal stressors and academic stressors. No gender differences were found between males' and females' use of problem-focused strategies for academic stressors. Adaptive problem-focused coping strategies were found to be more fully developed in older adolescents, whereas emotion-focused coping strategies were not as well-developed in these age groups. Academic stressors were perceived as more controllable than social stressors and more problem-focused alternatives for coping were used with academic stressors. Stvle Adolescent coping refers to specific behaviors used to manage demands of a particular situation to reduce or eliminate stress. A pattern of coping responses or a generalized strategy of approaching problems is usually developed that portrays a certain style of coping. Adolescents develop patterns of behaviors to deal with a multiple set of demands that may emanate from their roles as students, children, friends, and life events such as death, moving, divorce, and illness. These styles are developed

PAGE 56

46 depending on previous personal experience with similar situations, vicarious experiences associated with observing the success or failure of others, perceptions of personal vulnerability, and social persuasion by parents, peers, and significant others (Patterson & McCubbin 1987; Werner & Smith, 1982) . Coping styles emerge during adolescence that will have long-term consequences on coping styles used as adults (Valliant, 1977) . Patterson and McCubbin (1987) found 12 patterns of coping behaviors during their development of the instrument. Adolescent Coping Orientation for Problem Experiences (ACOPE) . The study was conducted with 4 67 junior and senior high school students from a suburban Midwestern school district. Patterns of behaviors were factor analyzed and grouped into the following categories: ventilating feelings, seeking diversions, developing self-reliance, developing social support, solving family problems, avoiding problems, seeking spiritual support, investing in close friends, seeking professional support, engaging in demanding activity, being humorous, and relaxing. Females were found to have higher mean scores for developing social support, solving family problems, investing in close friends and developing self-reliance. Males had higher mean scores than females on behavior patterns of being humorous. Females were found to use coping behavior more frequently directed at developing social support than males and both genders rated relaxing most frequently as a way to manage tension.

PAGE 57

47 Family-Based Coping Resources The family is viewed from a general systems theory (Bertalanf fy, 193 6) where the members and their relationships bind the system together. Every part of the system is related to its member parts so that change in one will cause change in the total system. Olson and McCubbin (1983) have delineated coping strategies and resources for functional and dysfunctional systems where 16 types of marital and family systems are based on the family dimensions of "cohesion" and "adaptability." This model of family functioning proposes that a balance of these two characteristics is the most functional to marital and family development. Parent-Adolescent Communication Clinical literature on childhood bereavement stresses that open communication between parents and children greatly facilitates the mourning process, and children are more likely to achieve a healthy adaptation to parental death when family relationships are characterized by open communication (Bowlby, 1963, 1980a, 1980b; Brice, 1982; Brown, 1988; Caine & Fast, 1966; Furman, R.A. , 1964; Furman, E., 1974, 1983; Kliman, 1979; Miller, 1971; Nagera, 1970). Brown (1988) described "openness" as the ability of each family member to stay nonreactive to the emotional intensity m the system and to communicate feelings to others without expecting others to act on them. The ability of the family to remain open, to express thoughts and feelings about the

PAGE 58

48 death, and to remain nonreactive to the anxiety of other family members is related to the intensity and duration of stress. Brown (1988) found that the longer and more intense the family stress is, the greater chance of pathological stress response. In a study of an adolescent son whose father died of cancer, lack of communication in the family caused isolation and anxiety that was expressed in physical symptoms leading to frequent hospitalizations of several family members. Children's adjustment is less difficult in families where open expression of anger, guilt, sadness, and loss is encouraged (Vollman, Ganzert, Picker, & Williams, 1971) The general level of communication established in early childhood influences a child's ability to talk about any topic and eventual attitudes toward life and death (Kastenbaum, 1986) . Open communication is of particular importance to stress management in families because it enables the members to coordinate their efforts in managing demands and helps to reduce ambiguity (Patterson, 1988). McGoldrick (1991) emphasized the importance of clear and open communication about the death and suggested that participation in funerals and graveside services are helpful to family members in integrating loss. Saler and Skolnick (1992) conducted a study of adults whose parent died in childhood and found that their ability to talk freely with the surviving parent

PAGE 59

49 about the circumstances around the death, express sorrow to the surviving parent, or ask questions about the deceased parent was protective against adulthood depression. The ability of children to talk about their feelings surrounding death helps them process grief (Moody & Moody, 1991; Segal, 1984; Silverman & Silverman, 1979) and discourages suppression of feelings which might lead to prolonged grieving. In a study by McNeil (1986), 335 adolescents from 12 to 18 years of age gave their perceptions of family communication patterns. The majority of this sample reported that they disclosed more to their peers before their parents on items relating to feelings about themselves, personal worries, and attitudes and opinions on love, sex, school, and friendships. However, students preferred to disclose their feelings and thoughts about the meaning of life and death and what happens after death to their mothers more than to their peers or fathers. Of the sample, 3 0% expressed a desire to talk more about deathrelated subjects with their parents than they had in the past. Goodman (1986) studied 30 adolescents following the death of a parent to observe grief reactions of normal adolescents. Adolescents reported that talking with peers and with others who experienced parental death was helpful in resolving grief but found it difficult to talk to those who had not had this similar experience. Similarly, Cragg

PAGE 60

50 and Berman (1990) in a study of 10 adolescents whose parent died of cancer found that adolescents fear being different from their peer groups and that it was difficult for them to turn to peers who had not experienced the loss of a parent. They also found that communication decreased between the surviving parent and adolescents after the death and adolescents were reluctant to discuss their feelings in order to protect the surviving parent from pain. Problems occur when the surviving parent looks to the children for extraordinary support after the death of a spouse. Children are unable to grieve until their parent has stabilized (Barth, 1989) . Family Adaptability Family adaptability is defined as the ability of the family system to change its power structure of roles and rules in response to situational and developmental stress (Gehring & Feldman, 1988). Barth (1989) interviewed a number of therapists and family members after the death of a parent and found that when problems arise, they are generally noticed 18 months to 2 years after the death. Immediately after the death, the family is preoccupied with soothing one another and not taking care of individual needs. Maladaption may not surface for years and may not emerge until another developmental stage is reached. The family's strengths and resources before parental loss are primary factors in the ability of the family to adjust. Connections with extended family and social support networks

PAGE 61

51 before death are also important to adaptation. Their visits and support provide a sense of continuity and safety (Barth, 1989) . Elizur and Kaffman (1983) found that the greatest threat to family adaptability was the surviving parent's inability to express grief, share the child's grief, and incorporate the expression of grief in the family system. Patterson and McCubbin's (1987) Double ABCX Model of Family Adaptation describes development, adaptation, and the influence of demands, resources, meaning, and coping on adolescents. The adolescent, as an individual, is viewed as part of a system together with family and community where reciprocal relationships are formed and the demands of one part of the system are met by the capabilities of another part. There is a "fit" with each system interface — individual-to-family, family-to-community, and individualto-community. Adaptation is achieved from the adolescent's perspective when there is a fit within the family and within the community of peers and in the school system (Patterson & McCubbin, 1987) . Patterson (1988) described a crisis as a state of diseguilibrium when a number of demands exceed the existing capabilities of the family and this imbalance persists. The family reaction during the adaptation phase is to restore balance or normalcy to the family by acquiring new adaptive resources and coping behaviors, reducing demands they must deal with, and changing the way they view their situation.

PAGE 62

52 Failure to make these changes may cause an increase in demands and cause the family to be more vulnerable to subsequent stressors. Davies et al. (1986) studied levels of family functioning in 111 families where a child had died of cancer from 2 to 9 years earlier. Rigidly maintained roles were found in less functional grieving families. The oldest child was assigned the role of the deceased child by the parents in some families, and in some families the child self-imposed the role. Less functional families were found to be resistant to reorganization and continued to function as if nothing had happened. They were unwilling to make new friends and were unable to mobilize resources. More functional families acknowledged their loss, reorganized and shared responsibilities, and were open, flexible, and adaptive in their reorganization and mobilization of resources. The Circumplex Model of the family system by Olson (1989) categorized families according to both their cohesion and adaptability scores. There were 16 subtypes depending on their degree of connection and flexibility. Olson and Stewart (1991) further suggested that there may be a relationship between these subtypes and the effects of stress. "Flexibly connected" families were shown to be more vulnerable to the accumulation of stressful events and "structurally separated" families were more affected by transitional changes. They suggested that life events and

PAGE 63

53 transitions are stressful to both types of families and stress is influenced by cohesion and adaptability. Family Cohesion Family cohesion refers to the emotional closeness that binds family members together and is the degree to which family members are separated from or connected to their family (Olson et al., 1992). Cohesion is generally linearly associated with positive adaptational outcomes in adolescent families. As familial roles and relationships are transformed throughout development, cohesion between parents and children is transformed from relatively high dependency on parents to a balanced connectedness permitting autonomy in adolescence (Noller & Callan, 1986) . Power relations also undergo transformations where parental unilateral power decreases and mutual power strategies by adolescents and parents increase (Youniss & Smollar, 1985). Barth (1989) found that the degree of closeness children have with parents is a determining factor in how they manage the loss of a parent. In studies of parentally bereaved children, Barth found that a child who is especially close to the surviving parent adjusts more easily to the death of the other parent. Conversely, one who is especially close to the parent who died, and distant from the surviving parent, has more difficulty adjusting. Elizur and Kaffman (1982) studied groups of bereaved kibbutz and non-kibbutz children at 18-month and 40-month postloss periods. The authors associated several child,

PAGE 64

54 family, and circumstantial variables in nonkibbutz children to suggest poor outcomes of adjustment including poor preloss adjustment, prior marital conflict, poor prior relationships between the child and the deceased parent, poor relations with the surviving parent, and the absence of a surrogate father. Wolfenstein (1966) theorized that when a parent dies during adolescence, the normal developmental processes of detachment may be avoided by overidealizing the deceased parent. Ambivalence is thought to be created through the overidealizing-detaching conflict and may cause hostility toward the surviving parent. The surviving parent may be blamed for the death and the adolescent may wish the surviving parent had died instead of the deceased. Alternatively, Wass & Stillion (1988) suggested that if other family members idealize the deceased parent and the adolescent has been successful in distancing and separating from the deceased, the adolescent may feel cut off and isolated from the other family members and feel misunderstood. In either case, considerable guilt may be generated and behavior problems may become evident. Males have been found to act out by stealing, using drugs, fighting, or demonstrating social withdrawal to escape pain. Whereas, girls were found to sexualize relationships and seek closeness in order to comfort themselves and replace their loss (Osterweis, Solomon, & Green, 1984).

PAGE 65

55 Silverman and Silverman (1979) found that it was easier for bereaved children to accept the loss of a parent if they could maintain a sense of connection to the parent in current life. Silverman and Worden (1992) studied the reactions of a nonclinical group of 125 children from 6 to 17 years of age who had lost a parent. They found that many of the children tried to stay connected to the parent through dreams, talking to or thinking about the deceased parent, claiming that the parent was watching them, visiting the parent's grave, and keeping things that belonged to the parent. The subjects tended to carry on a relationship with the deceased parent similar to the one prevailing before death. In a retrospective study of college-aged women who had lost a parent, the women were constantly renegotiating their relationship with their deceased parent. It was thought that by renegotiating a relationship with the deceased parent that it was easier to accept a new reality without the deceased (Silverman et al., 1992). Conclusion The death of a parent is believed to be the greatest stressor for an adolescent and family. Researchers have delineated normal and pathological responses to stress and have identified coping mechanisms used to adjust to changes created through loss. Developmental theorists have postulated that cognitive maturity influences coping styles,

PAGE 66

56 and life event theorists suggest that life events prepare individuals to cope with subsequent events. Numerous variables have been identified in the literature that are important to adolescent adjustment to parental loss, which include cognitive appraisal and decision making, coping strategies, personal resources, parent-adolescent communication, family adaptability, and family cohesion. Very few definitive studies exist, however, relating the adolescent's use of these resources to the stress resulting from parental loss. This study applies a transactional model of stress to assess the effects of adolescent-environment interactions. Stress response to parental death is measured and the influence of personal and family resources, and coping strategies are assessed. If significant relationships between the levels of stress reported by adolescents and the use of coping and family resources exist, then appropriate interventions might be designed to assist adolescents in developing these personal and family resources.

PAGE 67

CHAPTER 3 METHODOLOGY Statement of Purpose This study was designed to determine the degree of stress experienced by adolescents as a result of parental death and the relationships among personal and family coping resources and adolescent stress. The family coping resources assessed were family adaptability and cohesion and parent-adolescent communication. In this chapter the methodology used to address these issues is discussed. The chapter includes a description of the research design, the population and sample, the sampling procedures, the instruments used, the data collection procedures, and the data analysis procedures. Research Design A descriptive, correlational design was used with a gualitative component. The dependent variable in this study was adolescent stress response. Four independent variables used to predict the reported degree of stress were family adaptability, family cohesion, parent-adolescent communication, and types of adolescent coping strategies utilized. An open-ended guestionnaire inviting adolescents to describe the coping strategies they used with the death of 57

PAGE 68

58 the parent and a questionnaire requesting "expert advice" from the adolescent about what would be helpful to a friend who had experienced a similar death comprised the qualitative component of this research study. Delineation of Relevant Variables Dependent Variable The dependent variable is stress response . Subjective distress related to the specific life event of parental death was defined in terms of reported degree of stress. The Impact of Event Scale (lES) (Horowitz et al., 1979) was used to assess the current degree of stress. The scale is comprised of two response sets measuring intrusion and avoidance that when summed, determine the total subjective stress score. Independent Variable s Familv adaptability. Adaptability is the ability of the family system to change its power structure, role relationships, and relationship rules in response to situational and developmental stress (Olson et al., 1992). There is an increase in demands placed on the family after parental death. Death and transitions are stressful to families (Olson & Stewart, 1991) and adaptive resources help reduce the demands family members must deal with after the death of a parent (Patterson, 1988). In this study, family adaptability was measured by the Family Adaptability and Cohesion Scale (FACES II) (Olson et al., 1992).

PAGE 69

59 Family cohesion . Cohesion is defined as the extent of the emotional bonding that family members have toward one another. The degree of closeness children have with the surviving parent determines the ability of the child to adjust to the loss of a parent (Barth, 1989) . In this study, cohesion was measured by the Family Adaptability and Cohesion Scale (FACES II) (Olson et al. , 1992). Parent -Adolescent communication . Parent-adolescent communication was defined as the extent to which the adolescent shares factual and emotional information with the parent (Barnes & Olson, 1985) . The ability of family members to express thoughts and feelings about the death of a parent is related to the intensity and duration of stress (Brown, 1988) . Open communication between parents and children has been shown to facilitate the grieving process and promote healthy adaptation to parental death (Bowlby, 1980a, 1980b; Brice, 1982; Brown, 1988; Caine & Fast, 1966; Furman, 1983; McGoldrick, 1991; Moody & Moody, 1991; Segal, 1984; Silverman & Silverman, 1979). In this study, parentadolescent communication was measured by the ParentAdolescent Communication Scale (PAC) (Barnes & Olson, 1985). Adolescent copina. Coping was defined as the ability of the adolescent to regulate emotional distress and actively shape subsequent emotion (Folkman & Lazarus, 1980). Adolescent coping varies according to the stage of development, appraisal of the stressor, and the coping style (Patterson & McCubbin, 1987) . Patterns of adolescent

PAGE 70

60 behavior portray certain styles when approaching problems and are considered to be problem-focused or emotion-focused coping (Compas, et al., 1988; Patterson & McCubbin, 1991). Coping strategies are used to modify the stressor or to regulate emotional states that accompany stress (Folkman & Lazarus, 1980; Menaghan, 1983). Adolescent coping was assessed by the Adolescent Coping Scale (A-COPE) (Patterson & McCubbin, 1991) . Coping styles were identified by the frequencies of twelve descriptive patterns of behavior. Description of the Population The population for this study consisted of adolescents who had lost a parent through death within the past 30 months, were placed in regular classes in school (as opposed to special classes due to emotional or learning disabilities) , and were referred by staff members of the Hospice of North Central Florida, school counselors, or by members of the community to the researcher. The Hospice patient population was drawn from an 11-county area of North Central Florida encompassing Alachua, Bradford, Clay, Columbia, Dixie, Gilchrist, Hamilton, Levy, Putnam, Suwannee, and Union counties. The population in the State of Florida by race in 1990 was 84% White, 14% African American, 1% Hispanic, and 1% other (U.S. Bureau of the Census, 1990). The population of the 11 counties served by the Hospice of North Central Florida in 1990 was 80% White, 18% African American, and 2% other minorities (Bureau of Economic and Business Research,

PAGE 71

61 1991) . The population of Hospice families treated in 19941995 was 84% White, 14% African American, 1% Hispanic, and 0.3% Other. Table 3-1 shows the percentages by categories for each population. Table 3-1 Comparative Breakdown by Ethnicity for Population Subsamples African Population Subsample White American Hispanic Other Population of Florida, 1990 84% 14% 1% i% Population of Hospice of 84% 14% 1% 0.3% North Central Florida, 1990 Research Sample 58.6% 35.7% 3% 1% Sampling Procedures Applications to conduct research in public schools were filed in each school district in the 11-county area. The superintendent of schools in each county was mailed a letter (Appendix A) describing the study, together with copies of each instrument used in the study and a copy of a letter to school guidance counselors (Appendix A) . A follow-up phone call was made to the superintendent to answer any questions about the project and to determine if approval of the study was granted. The school boards in seven counties agreed to participate and four declined. Upon approval from the superintendent, principals and/or guidance counselors of the middle schools and high schools in each county were contacted and letters were sent to the

PAGE 72

62 guidance counselors. The guidance counselors identified students with parental loss and contacted the families to obtain permission to provide names, addresses, and phone numbers for contact by the investigator. Letters were then mailed to surviving parents and legal guardians describing the study (Appendix B) including a letter to the potential adolescent participant (Appendix C) . Follow-up phone calls were then placed to the parent/guardians and appointments were scheduled with students who were interested in participating in the study. Arrangements were made to meet the adolescents either at home or at school to complete the questionnaires . Other referrals were received from Hospice staff members and the coordinator of the Hospice Children's Program. Letters were mailed to the parents along with a letter to the adolescent explaining the research project. Follow-up phone calls were then made to the parent and upon approval of the parent, an appointment was scheduled to meet with the adolescents at their homes or at school. Ninety-eight letters were mailed to parents and adolescents and 70 agreed to participate in the study. Letters were first mailed on March 16, 1996, and all data was collected by May 11, 1996. There was a 71% return rate on requests to participate. The established minimum of 70 subjects were contacted and no additional procedures for obtaining subjects were instituted. All participants were given five dollars for participating in the study.

PAGE 73

63 Upon completion of the project, a letter was mailed to all subjects (Appendix D) thanking them for their participation that included a summary of coping skills reported by participants in this study to be helpful after parental loss. Sample The sample consisted of 70 adolescents between the ages of 12 and 19 (Table 3-2). Sixty-nine adolescents were between the ages of 12 and 18, and one 19 year old who was a senior in high school. There were 38 females (54.3%) and 32 males (45.7%) from 7 counties (Alachua [60.0%], Clay [7.14%], Columbia [1.43%], Dixie [1.43%], Levy [7.14%] Putnam [21.43%], and Suwannee [1.43%]). Ethnicity was represented by 41 Caucasians (58.6%), 25 African Americans (35.7%), 3 Hispanics (4.3%), and 1 other minority (1.4%). There were 26 maternal deaths (37.1%), 40 paternal deaths (57.1%), i grandmother (1.4%), and 3 grandfather (4.3%) deaths. Most of the deaths occurred within the past year. Frequencies of deaths included 20 deaths in 1994 (28.6%), 36 in 1995 (51.4%), and 14 within the past 5 months 1996 (20%) . The causes of death were described as follows: 47 due to a long illness (67.1%), 9 due to a sudden illness (12.9), 8 due to an accident (11.4%), 3 murders (4.3%), 1 suicide (1.4%), and 2 other (2.9%). Frequency distributions by age and county are shown m Tables 3-2 and 3-3. Frequency distributions by gender of

PAGE 74

64 deceased parent/guardian are shown in Table 3-4. Frequency distributions by type of death are shown in Table 3-5. Table 3-2 Frequency Distribution of Participant Age Total Sample: N = 70 Mean=14.2 SD=1.85 Age N Percent 12 17 24.3 13 11 15.7 14 13 18.6 15 13 18.6 16 8 11.4 17 3 4.3 18 4 5.7 19 1 1.4 Table 3-3 Breakdown by County for Research Sample County Research Sample Percent Alachua Brandf ord 42 60.00 Clay Columbia Dixie Gilchrist 5 1 1 7.14 1.43 1.43 Hamilton Levy Putnam 5 15 7.14 21.43 1.43 Suwannee Union 1

PAGE 75

65 Table 3-4 Frequency Distributions of Parental Death by Sex Sex of deceased Percent of Sample Maternal death 37.1 Paternal death 57. 1 Grandmother death 1.4 Grandfather death 4.3 Table 3-5 Frequency Distributions of Parental Death by Type of Death Cause of Death Research Sample Percent Long illness 47 67.1% Sudden illness 9 12.9% Accident 8 11.4% Murder 3 4.3% Suicide 1 1.4% Other 2 2.9% Data Collection Each adolescent was given a separate collection of questionnaires, coded in the upper right hand corner with a number (i.e., 101) and was asked not to put their names on the packets. Each participant completed a Demographic Questionnaire, Impact of Event Scale, FACES II, ParentAdolescent Communication Inventory, A-COPE, and Adolescent Questionnaire. There were a total of 128 questions and 2 short-answer questionnaires. Participants completed all questionnaires in 20 to 45 minutes, after which they received five dollars from the investigator for participation in the study. The investigator was present to

PAGE 76

66 answer any questions during the sessions and to inform the students of their eligibility to attend bereavement support groups and bereavement camps sponsored by the Hospice of North Central Florida. Instrumentation Six instruments were used in this study. Four standardized instruments were used as follows: (a) the Impact of Events Scale (IBS) , (b) the Family Adaptability and Cohesion Scales (FACES) , (c) the Parent -Adolescent Communication Scale (PACS) , and (d) the Adolescent Coping Orientation for Problem Experiences (A-COPE) . Two were investigator-developed and included the following: (a) a demographic questionnaire to determine information about the death of the parent and (b) an adolescent questionnaire to determine personal coping skills that adolescents found helpful in dealing with parental death and advice they would give others after similar losses. Measur es of Demographic Information The Demographic Questionnaire (Appendix E) was used to assess the age, gender, ethnicity of the adolescent, county of residence, relation to the deceased, age of the deceased, length of time since the death, and the type of death. Impact of Event Scale fIFS) The dependent variable was measured by the Impact of Event Scale (lES) developed by Horowitz et al. (1979). it was used to measure subjective distress or internal stress

PAGE 77

67 perceived by the adolescent due to the specific life event of the death of a parent. Subjective distress in response to life events is theoretically composed of two independent aspects of felt experience: intrusion and avoidance. Intrusion represents the involuntary awareness of ideas, memories, and emotions associated with the event. Avoidance represents the conscious attempt to divert attention away from cognitions and feelings related to the death. The lES includes response sets for both intrusion and avoidance with a sum score for a total measurement of stress. The scale is comprised of a 15-item self-report measure where 7 items describe episodes of intrusive thoughts and 8 items describe episodes of avoidance. The life event measured — the death of a parent — was listed at the top of the instrument and was the referent for each statement in the scale. The participant was asked to estimate the frequency to which each item described personal experience over the past week on a Likert 4-point scale ranging from "not at all" to "often." A score for each item was obtained by assigning weights of 0 for "not at all," 1 for "rarely," 3 for "sometimes," and 5 for "often." The intrusion subset of questions was represented by items 1, 4, 5, 6, 10, 11, and 14. The avoidance subset of questions is represented by items 2, 3, 7, 8, 9, 12, 13, and 15. An overall sum score presented the degree of current stress in relation to the death of the parent.

PAGE 78

68 The test was originally normed on a population of 66 adults who sought psychotherapy as a result of serious life events — half of which were bereaved and the other half who had experienced personal injuries. A cluster analysis was conducted on 20 original items in the test with a primary cluster comprised of clinically derived intrusion items and a second cluster comprised of avoidance items. The test was revised to include 15 items representing the 2 clusters. A split half reliability of the scale was high with r=0.86. Internal consistency of subscales using Cronbach's alpha was 0.78 for intrusion and 0.82 for avoidance. The correlation of 0.42 (p 0.0002) was found between intrusion and avoidance subscale scores, indicating the two subscales were associated but did not measure identical dimensions. Another population of 3 5 subjects who were out patients seeking psychotherapy after the death of a parent were compared to 37 field subject volunteers who experienced the death of a parent. All items were endorsed by both groups. Confirmation of validity of item assignments to intrusion and avoidance subscales was obtained in a factor analysis of the combined data. Subscales were internally consistent in both patient and field groups as well as across time with a Cronbach alpha ranging from 0.79 to 0.92 (Horowitz et al., 1984). In another study, nonclinical parentally bereaved adolescents were studied over a period of 1 year and evaluations were made at 6 weeks, 7 months, and 13 months

PAGE 79

69 following parental death. Stress response symptomatology was measured by the Impact of Event Scale and scores were compared to a study of parentally bereaved adult children. All subjects reported moderate to high levels of intrusive and avoidant stress-related symptomatology initially with decreased symptoms over the year. More than half of the subjects continued to report moderate or high levels of distress 1 year later. Adolescents reported an equal or slightly higher sustained degree of stress compared with parentally bereaved adult patients or control subjects measured by the lES (Harris, 1991). Family Adaptability and Cohesion Scales (FACES-II) The independent variables of family cohesion and adaptability were measured by FACES II. The scale was originally developed by Olson (1989) as an outgrowth of their Circumplex Model which was a systematic assessment of families and their ability to cope with stress. The instrument was originally developed with the intent to be used with children and those with limited reading ability. During the development of the scale, a 50-item scale was administered to 2,412 individuals in a national survey and on the basis of factor analysis and reliability analysis, the scale was reduced to 30 items. The final scale contained 16 cohesion items and 14 adaptability items. The cohesion dimension contains 2 items for each of the following eight concepts: emotional bonding, family boundaries, coalitions, time space, friends, decision-

PAGE 80

70 making, and interests and recreation. There are four levels of family cohesion ranging from extreme low cohesion to extreme high cohesion and families are characterized as disengaged, separated, connected, and enmeshed. The balanced levels of family cohesion are known as separated and connected. The adaptability dimension contains 2 to 3 items for the following 6 concepts: assertiveness, leadership, discipline, negotiations, roles, and rules. The four levels of family adaptability range from extreme low adaptability or change to extreme high adaptability or change. The levels are characterized as rigid, structured, flexible, and chaotic. The balanced levels are known as flexible and structured. Balanced levels are generally viewed as being characteristic of healthy family functioning and extreme levels are seen as more problematic for families over time. Internal consistency reliability for FACES II was initially derived from a study with a sample of 124 university and high school students with an average age of 19.2 years. Reliability was .87 for cohesion, .78 for adaptability, and .90 for the total scale. Test-retest reliability measures were taken from the-50 item version and the time lapse between the first and second administration for the test was 4 to 5 weeks. The Pearson correlation for the scale was .84 (.83 for cohesion and .80 for adaptability) .

PAGE 81

71 The FACES II was administered to each participant. The participant was asked to read the statements and determine how frequently, on a scale from 1 "almost never" to 5 "almost always" the described behavior occurs in the family. Linear scoring suggests that high scores on adaptability and cohesion dimensions indicate "very connected" and "very flexible" measures. After cohesion and adaptability scores were obtained, corresponding scores for family type were interpreted through the linear scoring and interpretation chart by Olson et al. (1992). Family types were described as balanced, moderately balanced, mid-range, and extreme. Parent-Adolescent Communication Inventorv (PACI^ The independent variable, parent-adolescent communication, was measured by the Parent -Adolescent Communication Inventory (PACI) developed by Barnes and Olson (1985). PACI is a self-report instrument designed to measure both positive and negative aspects of communication as well as aspects of the content and process of parentadolescent interactions. The scale was initially normed on a sample of 433 students with the majority in the late adolescent age range of 16 to 20 years of age. There were 127 high school students and 306 college students in the study. Thirty-five items were selected from a review of the literature on parent-adolescent communication and other items were generated to measure different aspects of parent-

PAGE 82

72 teen interaction. Items were selected to measure both process and content of communication and provided a readable scale for youth as young as 12 years of age. Data from the pilot study were analyzed using factor analysis methods and 3 subscales were identified as open family communication, problems in family communication, and selective family communication. The scale was reduced to 20 items based on factor analysis. The scale was then administered to 124 subjects with a mean age of 19.2 years to determine the test-retest reliability with intervals of 4 to 5 weeks. Subsequent studies were conducted with adolescents and a final scale was then developed to consider both adolescent and parent views. The scale was designed to measure both positive and negative aspects of communication as well as aspects of the content and process of the parentadolescent interactions. The 20-item Likert-type scale consists of 2 subscales. The first subscale, "open family communication," measures positive aspects of parentadolescent communication with a focus on free flowing exchange of information, lack of constraint, and degree of satisfaction in interactions. The second subscale, "problems in family communication," focuses on negative aspects such as hesitancy to share, negative styles of interaction, and caution in selectivity of what is shared. The final scale was administered to a larger study with a total sample of 1,842 subjects. Sample I contained 925 and Sample II contained 916 subjects. The alpha

PAGE 83

73 reliabilities were .87 for open family communication, .78 for problems in family communication, and .87 for the total score indicating that the two subscales and the total scale are reliable. Although adolescent responses regarding each parent were similar, the intergroup differences were significant enough to warrant the need for reporting different norms. Norms are reported for four different subdivisions: 1) fathers reporting an interaction with teens, 2) mothers' reports, 3) adolescents' reports regarding their mothers, and 4) adolescents' reports regarding their fathers. Participants were given the 20 statement inventory and asked to rate responses on a Likert-style 5-point response range from "strongly disagree" to "strongly agree" for a total sum score. A total sum score is used to reduce response bias. Items from the subscales are intermingled and it is necessary to distinguish items from the subscales for scoring. Items from the first subscale, which measure more positive aspects of parent-adolescent communication, are scored by adding the response choices in the Likert-type scale (l=strongly disagree; 2=moderately disagree; 3=neither agree nor disagree; 4=moderately agree; 5=strongly agree) . Items from the second subscale, denoting problems in family communication, are reversed in value. These items must be flipped in point value prior to summing the response values of the second scale with the first subscale for a sum total score of positive parent-adolescent communication.

PAGE 84

74 Adolescent Coping Orientation for Problem Exp eriences The independent variable, adolescent coping behavior, was measured by the adolescent coping orientation for problem experiences (A-COPE) developed by Patterson and McCubbin (1991) . Three adolescent population samples were used by the developers of the instrument. The first sample consisted of 3 0 10th, 11th, and 12th grade males and females. They were asked to describe what they do to manage hardships and to relieve discomfort for a) the most difficult stressor event personally experienced, b) the most difficult stressor event experienced by another family member, and c) difficult life changes and strains experienced in general. Ninety-five coping behavior items were generated which reflected the adolescent's developmental tasks of developing an identity and differentiating from the family while simultaneously maintaining appropriate linkages to it. Coping behaviors represented three primary functions of coping: a) coping by direct action to modify the situation by reducing demands and/ or increasing resources, b) coping by altering the meaning of experiences through perception and appraisal, and c) coping by managing the tension or stress experienced. Sample 2 consisted of 467 junior and senior high school students including 185 males and 241 females. The average age was 15.6 years. The 95 coping behavior items were presented and adolescents were asked to record how often they used each behavior when they faced difficulties

PAGE 85

75 or felt tense. The responses were factor analyzed to determine the underlying structure of A-COPE. Sample 3 consisted of 709 adolescents and their parents who were part of a 3-year longitudinal study designed to examine the influence of family variables on the use of substances by adolescents. Ages of adolescents ranged from 11 to 18 years old. The adolescents were asked to complete the A-COPE and an annual substance abuse questionnaire . A hierarchical approach to coping was then developed with A-COPE. Two general levels of abstraction were defined: (a) coping behaviors were operationally defined through 95 specific items and (b) specific coping patterns which were combinations of coping behaviors. Responses of 467 adolescents from sample 2 identified 95 coping behavior items of which 27 items were deleted. Repeated factor analyses resulted in a final set of 54 coping behavior items with factor loadings above .40 on twelve factors with eigenvalues of 1.0 or greater. The 12 factors accounted for 60.1 percent of the variance of the original correlation matrix. The factor structure of the coping behaviors in ACOPE with respective Cronbach alpha reliabilities on 12 factors are presented in the following twelve coping patterns: ventilating feelings (alpha=.75), seeking diversions (alpha=.75), developing self-reliance and optimism (alpha=.69), developing social support (alpha=.75), solving family problems (alpha=.71), avoiding problems

PAGE 86

76 (alpha=.71), seeking spiritual support (alpha=.72), investing in close friends (alpha=.76), seeking professional support (alpha=.50), engaging in demanding activity (alpha=.67), being humorous (alpha=.72), and relaxing (alpha=. 60) . Concurrent validity was examined in relationship to substance use in adolescents in sample 3 . Substance use was reported to have positive associations with coping patterns of ventilating feelings for males (r=.13 to .17), females (r=.ll to .17), and developing social support for males (r=.16 to .25), females (r=.13 to .18). Positive associations between friend support and substance use suggest that the socializing role of peers may influence adolescents to use substances. Coping patterns that had negative associations with substance use were solving family problems for males (r=-.l3), females (r=-.lO to -.21) and engaging in demanding activity for males (r=-.l2), females (^=-'1^ to -.18). Seeking spiritual support for females was negatively associated with substance use (r=-.ll to -.21). Other examples of coping patterns and decrease of substance use reflect the reliability and validity of A-COPE as an instrument for measuring adolescent coping with life problems . Coping behaviors cluster into coping patterns which appear to have a discrete function. For example, "ventilating feelings" is a coping pattern and the following behaviors are associated: get angry and yell at people,

PAGE 87

77 blame others for what's going wrong, say mean things to people, swear, let off steam by complaining to friends, and let off steam by complaining to family members. Behaviors associated with ventilating feelings and avoiding problems are considered undesirable or negative, whereas the remaining 10 patterns are considered transformational coping or positive types of coping behaviors. Five of the coping patterns (developing social support, solving family problems, seeking spiritual support, investing in close friendships and seeking professional support) involve talking to other people to discover solutions to problems and to increase social support and are considered to be "direct action" or problem-focused coping. However, these behaviors may also help reduce tension and may indirectly lead to altered meanings. The coping process appears to be multidimensional and behaviors may serve more than one function (solve problems and manage emotions) . There have been reports of gender differences in patterns of coping behaviors where females used coping behaviors directed at developing social support more frequently than males, and males used behaviors directed at being humorous more frequently than females (Patterson, 1988) . Functions of coping (Lazarus, 1966; Menaghan, 1983; Moos & Billings, 1982; Pearlin & Schooler, 1978) are described as (a) direct actions to reduce demands or increase resources through problem-focused coping, (b) altering the meaning through appraisal, and (c) managing

PAGE 88

78 tension through emotion-focused coping. Moriarty and Toussieng (1976) suggest that coping behavior may serve more than one function and that coping involves both a response to demands of external situations and a response to one's feelings about the situation. An adolescent coping score for A-COPE is simply a total sum score which represents positive coping patterns of behavior. The 54 -item Likert-style instrument is administered and the participant is asked to circle responses from 1 "never" to 5 "most of the time." The coping score is determined by summing the scores for each of the items. However, 9 items are scored with reversed values to ensure that all items are weighted in the same positive direction for both the analysis and interpretation of results. Adolescent Questionnaire (Appendix F) The adolescent questionnaire is a two-part questionnaire designed by the investigator as a qualitative component of the study. It was formulated to (a) identify coping behaviors used by the adolescents that were helpful after the death of their parents and coping that is used presently and (b) to solicit "expert advice" from the adolescents about what they would tell their best friends if they experienced a similar situation with the death of one of their parents. The first part of the questionnaire was designed to allow the adolescent to discuss coping experiences freely without a predetermined list of coping

PAGE 89

79 behaviors. The second part of the questionnaire was designed to allow the adolescents to reflect upon their losses and again decide what was helpful to them, but from different perspective. The intent was to put the adolescents in an "expert role" such that the advice they might give a friend may be different from what they experienced but important to the grieving process. H ypotheses The following null hypotheses were evaluated in this study: 1. After controlling for age and length of time since death, there is no relationship between adolescents' use of positive coping strategies and their reported degree of stress following the death of a parent. 2. After controlling for age and length of time since death, there is no relationship between adolescents' perceptions of the extent of positive parent-adolescent communication and their reported degree of stress following the death of a parent. 3. After controlling for age and length of time since death, there is no relationship between adolescents' perceptions of the degree of family cohesion and their reported degree of stress following the death of a parent. 4. After controlling for age and length of time since death, there is no relationship between adolescents' perceptions of the degree of family adaptability and their reported degree of stress following the death of a parent.

PAGE 90

80 Data Analysis Analyses of data were conducted using multiple regression to determine the relationships between adolescent stress and personal and family coping resources. Other regression analyses were used to assess the relationships between overall coping strategy and "coping patterns" of behavior. Means, standard deviations, and score ranges were computed to describe the sample in terms of intensity of stress experienced, types of coping strategies and patterns of coping styles utilized, degree of family cohesion, degree of family adaptability, and the degree of parent-adolescent communication . Adolescent questionnaires were analyzed and all coping mechanisms used by participants were categorized and tabulated regarding frequencies of coping skills considered helpful to adolescents after the death of a parent. Similar analyses were conducted regarding the advice adolescents would offer a friend who had lost a parent. Specific data and analytic procedures and results are described in Chapter 4.

PAGE 91

CHAPTER 4 DATA ANALYSIS AND RESULTS This study was designed to examine the degree of stress in adolescents as a result of parental death. A second purpose of the study was to explore the relationship between the use of personal coping strategies and adolescent stress. The third purpose was to explore the mediating effects of family adaptability, family cohesion, and parentadolescent communication on adolescent stress. In addition, a qualitative analysis was used to assess personal coping mechanisms reported by adolescents after the death of the parent and coping skills that continue to be helpful in managing their grief. Adolescents also were asked for the advice they would give their "best friend" after a similar experience of parental death. In this chapter, the results of the study will be presented as they pertain to each of the four null hypotheses. Next, the results of the qualitative analysis will be reported in terms of frequencies of coping mechanisms described and coping strategies recommended for friends. Four null hypotheses were evaluated in this study. Analyses of data were conducted using multiple regression with the SAS General Linear Model (GLM) to determine the 81

PAGE 92

82 relationships between the outcome measure of stress and the independent measures of coping, parent-adolescent communication, family cohesion, and family adaptability. Further regression analyses were conducted to determine the relationships between the 12 patterns of coping behaviors and gender of adolescent, relationship of deceased, age of deceased, length of time since death, communication, cohesion, and adaptability. Means, standard deviations, and score ranges were computed to describe the sample. For purposes of determining levels of statistical significance, the Type I error rate of .05 was established as the decision rule for all statistical tests. Hypothesis One The first null hypothesis addressed the relationship between adolescent use of positive coping strategies and subjective stress. Means and standard deviations of the total stress scale as well as individual subscales of intrusion and avoidance are reported in Table 4-1. In addition, the sample mean and standard deviation are reported for coping, communication, cohesion, and adaptability measures. As noted, participants scored a total mean score of 37.36 (SD=10.34, range 16-59) on the Impact of Event Scale. Scores are similar to a parentally bereaved adolescent sample (mean 39) reported by Harris (1991) and a sample of parentally bereaved neurotic adults (mean 3 9.5 (SD=17.12, range 0-69) reported by Horowitz et al. (1979).

PAGE 93

83 Table 4-1 Means, Standard Deviations, and Ranges of the Dependent Variable, Stress, and Independent Variables: Coping, Communication, Cohesion, and Adaptability Variable Mean SD Range STRESS 37. 36 10. 34 16-59 Intrusion 19. 16 5. 57 7-28 Avoidance 18. 29 6. 20 8-32 COPING 168. 05 26. 37 101-225 COMMUNICATION 64. 79 15. 16 20-91 COHESION 52. 73 12. 29 24-74 ADAPTABILITY 43. 34 9. 38 14-60 The subscale of intrusion (Table 4-1) had a mean score of 19.16 (SD=5.57, range 7-28) compared to a sample of neurotic adults who had a higher mean of 21.4 (SD=9.6, range 0-35) . The avoidance subscale (Table 4-1) of subjects in this study had a mean of 18.29 (SD=6.20, range 8-32) and was similar to the sample of neurotic adults with a mean of 18.2 (SD=10.8, range 0-38). Measurements of coping on the A-COPE (Table 4-1) produced a mean score of 168.05 (SD=26.37, range 101-225). This mean is similar to normed scores for Black adolescents (mean of 169.43, SD=25.23) and White adolescents (mean 167.86, SD=24.45) from single parent families (Patterson & McCubbin, 1991) . Null hypothesis one states that after controlling for age and length of time since death, there is no relationship

PAGE 94

84 between adolescents' use of positive coping strategies and their reported degree of stress following the death of a parent. A series of three multiple regression analyses were conducted. In the first multiple regression analysis in which the total stress score served as the dependent variable, there was no significant relationship between the independent variable, coping, and the dependent variable, the total stress score. Results of the analysis are shown in Table 4-2. In a second multiple regression analysis, in which the intrusion subscale score served as the dependent variable, no significant relationship was found between the variables (R-Square=0 . 09 , p=.77). In the third multiple regression analysis, in which the avoidance subscale score served as the dependent variable, no significance was found among variables (R-Square=. 06, p=.89). The model failed to achieve significance for the independent variable of coping (F=.52, p=.47) (Table 4-3). Therefore, no statistical evidence existed to reject the null hypothesis. Analyses of subscales are shown in Tables 4-3 and 4-4. Null Hypothesis Two The second null hypothesis addressed the relationship between positive parent-adolescent communication and subjective stress. A mean score of 64.79 was computed for

PAGE 95

85 Table 4-2 Hierarchial Regression Analysis of the Relationships Among Age, Length of Time Since Death, Communication, Cohesion, Adaptability with Total Stress Score as the Outcome Variable Total Sample N = 70 Variables df Type III SS F-value p-value Age 1 7.4749 0.07 0.78 Length 1 121. 5754 1.19 0.28 Cope 1 53.3848 0.52 0.47 Comm 1 24 . 3430 0.24 0. 62 Cohes 1 99.3632 0.97 0. 32 Adapt 1 32 . 4367 0. 32 0. 57 Variables Est SE t-value p-value Intercept 38.25 14.53 2.63 0.01 Age 0.22 0.81 0.27 0.78 Length -2.24 2.05 -1.09 0.28 Cope 0.04 0. 05 0.72 0.47 Comm -0. 07 0. 14 -0.49 0.62 Cohes -0. 18 0. 19 -0.99 0.32 Adapt 0. 14 0.26 0.56 0.57 R-Square=0 . 09 *p <.05

PAGE 96

86 Table 4-3 Hierarchial Regression Analysis of the Relationships Among Age, Length of Time Since Death, Communication, Cohesion, Adaptability with Intrusion Subscale Score as a Measure of Stress Variables df Type III SS Fvalue p-value 1 \J • UX^ / u . u u n o o 1 J 7 • ^ Q U ^ 1 TO U . 1 X c^R 9090 ^ O • 1 Q A U . Xo X Z> m O H O J n 1 Q U . o / 1 X X*t • J O X X n A Q U • 4o U . 4 y X n no U . / / Variables Est SE t-value p-value Intercept 17.28 7.80 2.21 0.03 Age 0.00 0.43 0.02 0.98 Length -1.24 1.08 -1.15 0.25 Cope 0.04 0.03 1.40 0. 16 Comm -0.03 0.07 -0.42 0. 67 Cohes -0. 07 0. 10 -0.69 0.49 Adapt 0.04 0. 14 0.29 0.77 R-Square=0. 09 * p <. 05

PAGE 97

87 Table 4-4 Hierarchial Regression Analysis of the Relationships Among Age, Length of Time Since Death, Communication, Cohesion, Adaptability with Avoidance Subscale Score as a Measure of Stress Variables df Type III SS F-value p-value An o X O Q R Q 1 r\ r\ "7 U • U / 0 . 78 ±J\H l\J L.Xi X A A C "7 Q £ 1 . 11 0.29 1 X r\ n n o 0.00 0 . 99 1 X T T Q 1 n J . jyiu 0 . 08 0.77 Cohes 1 42.5985 1.06 0.30 Adapt 1 14.8818 0.37 0.54 Variables Est SE t-value p-value Intercept 22.24 8.85 2.51 0.01 Age 0.13 0.48 0.27 0.78 Length -1.32 1.26 -1.05 0.29 Cope -0. 00 0.03 -0.01 0.99 Comm -0.02 0.09 -0.29 0.77 Cohes -0.12 0.12 -1.03 0.30 Adapt 0. 10 0.16 0.61 0.54 R-Square=0 . 06 * p <.05 this sample on the Parent-Adolescent Communication Inventory indicating an average sense of positive communication with surviving parents. This mean score is similar to both reported norms by Barnes and Olson (1985) for motheradolescent communication (mean 66.56, SD=12.10, range 20100) and for father-adolescent communication (mean 63.74, SD=12.02, range 20-100). The aspects of parent-adolescent communication assessed by the instrument were the amount of openness, the extent of problems or barriers of family

PAGE 98

88 communication, and the degree to which adolescents are selective in their discussion with parents. Null hypothesis two states that after controlling for age and length of time since death, there is no relationship between adolescents' perception of positive parentadolescent communication and their reported degree of stress following the death of a parent. The results of multiple regression analysis indicate that there was no significant relationship between parentadolescent communication and reported degree of stress. The model failed to denote significance between parentadolescent communication and stress (F=.24, p=.62) (Table 42). Therefore, the null hypothesis was not rejected. Null Hypothesis Three The third null hypothesis addressed the relationship between family cohesion and subjective stress. The mean score for family cohesion was 52.73 (SD=12.29, range 20-91). The linear scoring and interpretation procedure (Olson et al., 1992) indicates that the mean score for this sample denotes the average family as "separated" on the cohesion subscale of the FACES II. While this score falls within the balanced range of family types, it is slightly lower than an adolescent normed sample score of 56 reported by Olson and McCubbin (1983). This suggests that the score is consistent with studies of adolescents who typically report lower family cohesion because of their developmental stage and increased autonomy.

PAGE 99

89 Null hypothesis three states that after controlling for age and length of time since death, there is no relationship in adolescents' perceptions of the degree of family cohesion and their reported level of stress following the death of a parent. The results of multiple regression analysis indicate that there was no significant relationship between family cohesion and the reported degree of stress (F=.97, p=.32). The null hypothesis was not rejected. Results are shown in Table 4-2. Null Hypothesis Four The fourth null hypothesis addressed the relationship between family adaptability and subjective stress. A mean score for adaptability in this study (43.34, SD=9.38, range 14-60) was slightly lower than a normed sample of adolescents with a mean score of 46 reported by Olson and McCubbin (1983). The linear scoring and interpretation scale (Olson et al., 1992) was used to characterize the family as "structured" based on scores of the FACES II. The combined subscores for cohesion (52.73, "separated") and adaptability (43.34, "structured") indicated that the "structurally separated" scores produced an average family type description of "mid-range" families. Olson and Stewart (1991) reported that "structurally separated" or mid-range family types are not as vulnerable to stressful life events as much as extreme family types.

PAGE 100

90 Null hypothesis four states that after controlling for age and length of time since death, there is no relationship in adolescents' perceptions of the degree of family adaptability and their reported degree of stress following the death of a parent. The results of multiple regression analysis indicate that there were no significant effects for family adaptability and the reported degree of stress (F=.32, p=.57). Therefore, the null hypothesis was not rejected. Analyses are shown in Table 4-2. Adolescent Use of Coping Strategies While stress was the major focus in this study, an understanding of how participants utilized particular coping styles of behavior in managing stress was also important. A follow-up analysis was conducted to assess the frequencies with which certain coping styles were used by the sample and to assess the relationships between the twelve patterns of coping behaviors and the independent variables. Within the coping scale, there are distinct clusters of coping behaviors that represent certain styles of coping. These behaviors are classified on the basis of function, and different patterns may serve more than one function. Means and standard deviations for coping styles used by this sample are reported in Table 4-5.

PAGE 101

91 Table 4-5 Means, Standard Deviations, and Ranges of the Twelve Patterns of Coping Behavior Were Computed to Assess Coping Styles Used by Subjects in this Study Mean SD Range venuixauing reexings 0. 63 0 . 11 0 .33-0 . 83 oeeKing uiversions n ^ o 0 . 14 0 . 20-0 . 93 ueveioping oeir— Keiiance 0. 67 0 . 15 0 . 20-1. 00 and Optimism Developing Social Support 0. 66 0 . 15 0 .37-1. 00 Solving Family Problems 0. 60 0 . 18 0 .24-1. 00 Avoiding Problems 0. 80 0 .10 0 .48-0. 96 Seeking Spiritual Support 0. 50 0 .20 0 .20-1. 00 Investing in Close Friends 0. 68 0 .23 0 .20-1. 00 Seeking Professional Support 0. 42 0 .21 0 .20-1. 00 Engaging in Demanding 0. 67 0 .17 0 .30-1. 00 Activity Being Humorous 0. 68 0 .25 0 .20-1. 00 Relaxing 0. 69 0 .12 0 .45-0. 95 The most freguently used coping patterns (avoiding problems and relaxing) in this sample were normatively evaluated as undesirable by Maddi (1981). The next three most frequently selected coping patterns (investing in close friends, developing self-reliance, and engaging in demanding activity) are considered transformational coping styles using direct action to solve problems. The least frequently used coping pattern for this sample (seeking professional support) was also the least chosen by a normed sample of adolescents reported by Patterson and McCubbin (1991). Additional regression analyses were conducted to assess the relationships between 12 coping styles and the following predictors: gender of the adolescent (l=female, 2=male), relationship of the deceased (l=mother, 2=father, 3=grandmother, 4=grandf ather) , age of the deceased, length

PAGE 102

92 of time since the death, communication, cohesion, and adaptability. Significant relationships were shown between particular independent variables and the following described coping styles: engaging in demanding activity, seeking professional support, avoiding problems, solving family problems, developing social support, and seeking diversions. Results are shown in Tables 4-6 to 4-11. Enaaainq in demanding activity is a coping style that includes behaviors that challenge the adolescent to excel at something or achieve a goal such as strenuous physical activity, improving oneself, or working hard on school work (Patterson, 1991) . For this coping strategy, 18% of the variance (R-square=. 177) was accounted for by the model, (F(9,54)=1.29, p=.26). The Beta weights are presented in Table 4-6. As is evident in the table, cohesion (B=-.007, p=.02) was a significant predictor of this coping style. Low family cohesion was associated with a high use of engaging in demanding activity as a coping method. Seeking professional support focuses on behaviors directed at getting help and advice from a professional counselor or teacher about difficult problems. For this coping style, 13% of the variance (R-square=. 131) was accounted for by the model, (f(9,55)=.92, p=.51). The Beta weights are presented in Table 4-7. In contrast to the males, females reported that they sought out professional support at a significant level. Length of time since death

PAGE 103

93 Table 4-6 Beta Weights for Relationships Between Independent Variables and Coping by Engaging in Demanding Activity Variables Estimate p-value 4* ^* 4* xn tercep t 0 . 573 0 . 017 Gender 1 0.002 0.966 Gender 2 0.000 Deceas 1 0.152 0. 161 2 0.043 0.688 3 0.321 0. 135 4 0.000 Age 0.004 0.765 Length -.005 0.890 Comm 0. 003 0.170 Cohes -.007 0. 023 Adapt 0.003 0.480 R-Square=0. 177 *p <.05 Table 4-7 Beta Weights for Relationships Between Independent Variables and Coping by Seeking Professional Support Variables Estimates pvalue Intercept 0. 514 0.00 Gender 1 0.117 0.69 2 0.000 .05 Deceas 1 2 0.164 0. 154 0.23 3 4 0.169 0. 000 0.48 • Age Length Comm Cohes Adapt -.022 -.074 0.000 0.001 0.001 0. 17 0.08 0.40 0.02 0.61 R-square=0 . 34 ''P <.Ub — .

PAGE 104

was a negative predictor (B=-.074) and indicated that adolescents were more likely to seek professional support more frequently at times closer to the occurrence of death. Avoiding problems includes behaviors which involve the use of activities such as drinking beer or smoking as a way to escape or avoid persons or issues which cause problems, like staying away from home or telling oneself the problem is not important. A significant portion of the variance (R-square=. 343 ) in the coping style, avoiding family problems, was accounted for by the model, (F (9,54)=3.15, p=.004). The Beta weights are presented in Table 4-8. The significant predictor for use of this coping style was cohesion (B=.004, p=.03) indicating that the higher the level of family cohesion reported, the higher the frequency with which avoidance of family problems was reported as a coping strategy. Solving family problems includes behaviors that focus on open communication with family members, doing things together, and following family rules to minimize conflict. Fifty-two percent of the variance (R-square=. 523) in the coping style, solving family problems, was accounted for by the model, (F(9,53)=6.45, p=.0001). The Beta weights are presented in Table 4-9. Communication (B=.004, p=.04) was a positive predictor for the use of coping through solving family problems. Thus, females with higher communication were more likely to cope by solving family problems.

PAGE 105

95 Table 4-8 Beta Weights for Relationships Between Independent Variables and Coping by Avoiding Problems Variable Estimate p-value Intercept 0.637 0.000 Gender 1 2 0.010 0.000 0.694 • Deceas l 2 3 4 0. 063 0.097 0.073 0. 000 0.232 0.066 0.485 • Age Length Comm Cohes Adapt -.009 0.029 -.001 0.004 0.001 0.171 0.088 0.405 0.027 0.619 R-square=. 344 Table 4-9 Beta Weights and Coping by for Relationships Between Solving Family Problems Independent Variables Variables Estimate p-value Intercept Gender l 2 Deceas 1 2 3 4 Age Length Comm Cohes Adapt R-square=0. 52 -.037 0.832 0. 095 0.012 0. 000 • 0. 021 0.791 -.003 0.974 0. 095 0.552 0. 000 • 0. 002 0.872 0. 016 0. 523 0. 004 0.043 0. 004 0. 116 0. 002 0. 500 *p <.05

PAGE 106

96 Females (B=.095, p =.01) reported using this coping method at significantly higher levels than males. Developing social support includes behaviors directed at efforts to stay emotionally connected with other people through reciprocal problem solving and expression of affect. Examples of these behaviors are helping others solve their problems, talking to a friend about one's feelings, and apologizing to others. In the coping style of developing social support, 36% of the variance (R-square=. 36) was accounted for by the model (F (9,52)=3.18, p=.004). Beta weights are presented in Table 4-10. Females (B=.123, p=.003) reported using social support at significantly higher levels than males. Negative predictors were length of time since death (B=-.02, p=.44), communication (Beta=-.001, p=.7l), and cohesion (Beta=-.002, p=.46). This indicated that the shorter the length of time since death, the more adolescents used social support as a coping strategy. When parent-adolescent communication was less positive, coping through social support was higher. Social support also was used at a higher rate when the degree of family cohesion was low. Seeking diversions includes behaviors focused upon adolescent efforts to keep busy and engage in relatively sedate activities as a way to escape from or forget about the sources of tension and stress such as sleeping, watching TV, or reading. m the coping style of seeking diversions, 25% of the variance (R-sguare=. 25) was accounted for by the model, (F(9,54)=1.99, p=.058). The Beta weights are presented in Table 4-11. As is evident in Table 4-11,

PAGE 107

97 Table 4-10 Beta Weights for Relationships Between Independent Variables and Coping by Developing Social Support p— value Intercept 0.617 0. 001 Gender 1 0.123 0.003 n nnn U . \}\J\J • Deceas 1 0. 025 0.764 2 -.078 0.351 3 0.037 0.828 4 0.000 • Age 0.003 0.735 Length -.020 0.446 Comin -.001 0.714 Cohes -.002 0.523 Adapt 0.003 0.459 R-square=. 355 *p <.05 the significant predictors were cohesion (B=-.006, p=.02), and deceased mother (Beta=.l78, p=.03). Cohesion was a negative predictor and indicated that the more cohesion in families, the less adolescents used diversions as a coping method. When the mother was the deceased parent, adolescents reported greater use of diversions as a coping method than with the grandfather. Adolescent Questionnaire Participants were invited to discuss what they believed to be the most important things that were helpful to them during the time right after the death and things that remain helpful to them while coping with their loss. They were then asked to write a statement of advice they

PAGE 108

98 Table 4-11 Beta Weights for Relationships Between Independent Variables and Coping by Seeking Diversions V ar^XalJX6 ssdmate pvalue T ^% V* xnuercepu 0 . 558 0. 002 VjcllUcX X 0 . 001 0. 868 2 0.000 • Deceas 1 0. 177 0.025 2 0.098 0.204 U . 2 03 0. 189 4 0.000 Age -.006 0.542 Length -.042 0.090 Coitun 0.002 0.325 Cohes -.006 0.015 Adapt 0.005 0. 132 R-sguare=0 . 25 *p <.05 would give their best friend if the friend had a parental loss similar to the loss experienced by the participant. Responses were coded with key coping behaviors. Of the participants, 90% reported that being with and talking to friends and family about the loss was the most helpful coping mechanism used. Participants rated talking to friends (41%) with greater frequency than talking to parents or family (37%) and listed just "talking about it" (11%) as helpful. Adolescents advised friends to talk with "friends" with greater frequency (39%) than they advised their friends to talk with parents or family (14%) in the event that they experienced parental death. Crying (13%) was listed as the next best helpful coping mechanism, praying (10%), thinking of the good times

PAGE 109

99 spent with the deceased (10%) , not talking about it (7%) , and using sports (4%) as a way to cope with the loss were cited. Other coping activities reported by participants were sleeping, riding a bike, pounding a pillow, yelling, doing something fun, looking at pictures of the parent, punching, listening to music, writing letters, reading, going to the Hospice Kids Camp, being by "myself," talking to teachers, going to the batting cage, walking, and holding a stuffed animal. Adolescents gave the following "expert advice" to friends who might encounter a parental death: "talk to me" (23%), "talk to friends" (16%), have faith in God and pray (16%), be with and talk to a parent or family member (14%), think about the good times you had with the deceased parent (13%), and try not to think about it (4%). Other recommendations by participants were to keep some of the parent's belongings, believe that the parent is in a better place and not suffering any longer, help the surviving parent, believe that things will get better, take up a hobby, talk to the deceased, talk to a professional, think about the deceased parent, and play sports. Summary This chapter described the procedures for analyses and results of this research project. The outcome of findings was based on the decision to accept or reject the null hypotheses. The statistical evidence derived from analyses supported the rejection of all hypotheses, one through four.

PAGE 110

100 Regression analyses revealed that there was no support for significant associations between subjective stress and the independent variables of coping, communication, cohesion, and adaptability. However, there were significant relationships reported between specific types of coping behavior patterns and the gender, relationship to deceased, length of time since death, communication, and cohesion. Analysis of the responses to the adolescent questionnaire revealed that social support from friends and family was considered the most helpful coping mechanism for adolescents during their grieving process. Talking to a friend was the singular most important piece of advice adolescents recommended to a friend who might experience a similar parental death.

PAGE 111

CHAPTER 5 DISCUSSION The purpose of this study was threefold. First, the degree of stress reported by bereaved adolescents was examined. Second, the mediating influences of personal coping strategies and family life on adolescent stress response was explored. The three aspects of adolescent family life assessed were family adaptability, family cohesion, and parent-adolescent communication. Third, the types of coping methods used by adolescents and their recommendations to friends about coping strategies were assessed. In this chapter, a discussion of the results of the research questions, the styles of coping, the limitations of the study, its implications, and suggestions for future research are presented. Description of Sample The study sample consisted of 70 adolescents in which 59% were Caucasian and 36% were African American with a mean age of 14.2. There were more females (54.3%) than males (45.7%). Paternal death (57.1%) was reported at a higher rate than maternal death (37.1%) and grandparent deaths (5.7%). The majority of deaths were caused by long-term illnesses (67.1%), and most of the other deaths were caused by sudden illnesses and accidents (24.3%). All deaths occurred within the past 30 months with an average length of time since death of 1 year. 101

PAGE 112

102 The degree of stress found in adolescents in this study was moderate to high as compared with other populations, and transformational coping behaviors were used by the majority of adolescents with the exception of one emotion-based coping pattern of "avoidance." The average family type was "mid-range," indicating moderate levels of cohesion (separated) and adaptability (structured) . Communication between parents and adolescents was considered positive and open with the majority of participants. Question One: Relationship Between Stress and Coping Does the use of certain coping strategies relate to the reported degree of stress in parentally bereaved adolescents? This question examined whether coping methods directly influenced the degree of stress after the death of a parent. Results indicated there was no significant relationship between stress measures on the lES and coping measures on the A-COPE in this study. Stress scores derived by the subscales of the lES indicated similar means on intrusive thoughts and avoidance, suggesting that adolescents in this study experienced oscillations between episodes of intrusion and avoidance. An explanation for these findings is that adolescents may have incomplete personal schema for understanding death, as reported by Horowitz et al. (1993). In their completion tendency theory, Horowitz et al. posit that intrusive thoughts are cognitive attempts to assimilate stressful memories of the death experience into pre-existing schemata. In contrast, avoidance is seen as an attempt to protect the person from states of mind that are too painful or too

PAGE 113

103 emotional. The process is a mechanisin for maintaining control and safeguarding against hopelessness. The two most frequently chosen intrusion statements by participants were "my feelings about it were kind of numb" and "pictures about it popped into my mind." The two most frequently chosen avoidance statements were "I tried not to talk about it" and "I avoided letting myself get upset when I thought about it or was reminded of it . " The majority of coping strategies used by adolescents in this study were "problem-focused" such as investing in close friends, engaging in demanding activity, and developing social support as opposed to "emotion-focused" methods such as avoidance or yelling. The former is an attempt to modify the stressor through direct action and the latter is an attempt to regulate emotional states that accompany stress. Although there were no indices for deviant behavior problems or depression in this study, there were also no extreme scores found in family variables to indicate pathological response to stress. This observation is consistent with other studies where adolescents who used more problem-focused strategies had fewer emotional and behavioral problems (Compas, 1988; Ebata & Moos, 1991; Harris, 1991; Patterson & McCubbin, 1991). The coping style with the highest mean score was "avoiding problems" and is considered emotion-focused coping. Coping patterns directed at avoidance are normatively evaluated as undesirable and include ventilating feelings, seeking diversions, avoiding problems, and relaxing (Maddi, 1981). Interpretation of results in this

PAGE 114

104 Study are divergent and suggest that although avoidance may not be considered a positive mechanism, a certain degree of avoidance may be essential in assimilating the loss and helping in the adjustment of adolescents to the loss of a parent. There were no observed significant relationships between coping and gender in this study, contrary to Patterson & McCubbin's (1987) report that males favored coping through the use of humor and females favored developing social support as ways to manage tension. However, of those participants who used developing social support as a coping style, females reported higher scores than males. Question Two; The Relationship Between Stress and Parent-Adolescent Communication Does the degree of parent-adolescent communication influence the degree of stress reported by parentally bereaved adolescents? In the second question, the ability of the adolescent and parent to freely express thoughts and feelings while remaining nonreactive to one another was related to the degree of stress in the family after the death of a parent. Results of the analysis indicate there was no significant relationship between stress and parentadolescent communication. However, a number of participants reported difficulty in talking to parents about the death. This may suggest that adolescents were reluctant to discuss their feelings about death in order to protect the surviving parent from pain (Cragg & Berman, 1990) .

PAGE 115

105 On the other hand, more than half of the participants reported positive, open communication between the adolescent and the surviving parent. The most frequent responses of subjects were "my parent tries to understand my point of view," "my parent is a good listener," and "when I ask questions, I get honest answers from my parent." This data supports a similar study by McNeil (1984) where adolescents reported a desire to talk more about death-related subjects with their parents than with others. Reports of positive communication with parents is consistent with the qualitative analysis of coping skills in this study. The most frequently rated coping strategy was talking to friends and family members about the death of the parent. Question Three; Relati onship Between Stress and Cohesion Does the degree of family cohesion influence the degree of stress reported by parentally bereaved adolescents? In the third research question, the adolescent's perception of family influences such as emotional bonding, boundaries, coalitions, decision-making, and recreation was related to the degree of reported stress after parental death. The results of the analysis indicate there was no significant relationship between stress and family cohesion. Cohesion was rated according to the Linear Scoring Scale of FACES-II with a range from very connected, connected, separated, to disengaged. The average rating of family cohesion by adolescents in this study was "separated."

PAGE 116

106 This rating may be explained by developmental studies of adolescents who have successfully separated from relationships of high dependency on parents to positions of independence as they establish autonomy (Noller & Callan, 1986; Wass & Stillion, 1988; Youniss & Smollar, 1985). These results imply that the average score on cohesion in this study may be confounded by the developmental stage of adolescence. In cases of complicated bereavement, Elizur and Kaffman (1983) suggested that low cohesion may be associated with poor preloss adjustment, prior marital conflict, poor prior relationships with the deceased parent, and poor relations with the surviving parent. Adolescents may feel cut-off and isolated from the surviving parent and the family. Alternatively, complicated bereavement may also be associated with high cohesion where the adolescent engages in compulsive caregiving of the surviving parent and is preoccupied with the death of the parent (Harris, 1991). Cohesion was a predictor of the two coping styles of seeking diversions and avoiding problems in this study. Behaviors characteristic of these styles were to escape and attempt to forget about sources of stress by sleeping, watching TV or reading, and avoiding persons or issues which cause problems or produce stress. Overall, adolescents reported functional families in which cohesion was positive and family type was on the lower end Of the scale of being balanced. Positive adaptational

PAGE 117

107 outcomes were inferred with the average family in this study (Noller & Callan, 1986) . Question Four; The Relationship Between Stress and Adaptability Does the degree of family adaptability influence the degree of stress in parentally bereaved adolescents? In the fourth research question, the ability of the family system to be flexible, reorganize, and share responsibilities was related to the degree of stress after the death of a parent. The results of analysis indicated there was no significant relationship between stress and family adaptability. The Linear Scoring Scale of FACES II rates family adaptability with a range from very flexible, flexible, structured, to rigid. The average rating of family adaptability is this study characterized families as "structured." A combined score of family cohesion and family adaptability indicated that the average family type in this sample is characterized as a "mid-range" family. Olson and Stewart (1991) suggested that "structurally separated" family types were less vulnerable to stressful life events and more affected by transitional change. Interpretation of the average family in this study suggests that families were adaptive in their reorganization after parental death and were able to mobilize the resources necessary to remain functional. Davies et al. (1986) found this type of grieving family capable of acknowledging their loss, reorganizing and sharing responsibilities, and adapting to their environment. In contrast, less functional families were found to maintain rigid roles, be resistant to

PAGE 118

108 reorganization, be unable to mobilize resources, and continue to function as if nothing had happened. In summary, the mid-range families in this study were described as functional and were on the lower end of the flexibility scale. These results may be influenced by the developmental stage of adolescents. Establishing an independent identity and separating from parents is complicated by the inherent life changes that occur with the death of a parent. Adolescent Questionnaire: Coping Behaviors and Advice In addition to standardized instruments, a qualitative analysis was conducted in which participants were asked to discuss coping methods that were helpful to them after the death of their parents. They were also asked to give the advice they would tell a friend who experienced the death of a parent with similar circumstances to the death of their parents. Participants clearly identified coping methods that were helpful to them. A high degree of family and social support was reported by participants, and 90% indicated they would advise friends to "talk" about their experience of parental loss with a friend or family member. The most frequent coping methods recommended by participants for managing grief were "talking," "crying," "praying," and "thinking about the good times." Some of these coping strategies are considered transformational coping methods because they imply direct action in discovering solutions to problems and increasing social support. Others involve reducing tension, which may

PAGE 119

109 lead to altered meanings attributed to the death. Responses to the questionnaire are similar to responses to the A-COPE and suggest that participants in this study used a higher percentage of problem-focused coping methods than emotionfocused methods. Limitations of the Study Descriptive studies, such as this one, have several limitations. Because the study was not longitudinal, the direction of causal effects among variables cannot be assessed. For example, it can not be determined if the degree of stress effected coping or coping effected stress. Further, the degree of stress in bereaved adolescents may be confounded with other stressors arising from sources not measured in this study. Stress is not a pure measurement and it is questionable about what is being measured. For example, stress could be confounded with anxiety. Another limitation is the sample selection procedure. Constraints were placed on selection because parental permission was necessary in order to contact participants. A sampling bias was evident because only those adolescents whose parents gave permission for their children to participate could be contacted and certain groups of adolescents may have been excluded. It may be that only those families who had a high level of family functioning were included in the study. There were also a number of adolescents who had previously attended bereavement group therapy sessions and bereavement camps.

PAGE 120

110 Thus, results from the study may not generalize to other groups of adolescents who have not received interventions. The self-report instruments used in the study require certain levels of cognitive abilities for comprehension, although the instruments have been normed for ages 12 and over. The Impact of Event Scale (lES) measured current subjective distress, and responses were restricted to feelings and thoughts that occurred "within the past seven days." The accuracy with which participants followed the directions and limited self -reports to the past 7 days was a concern. The Impact of Event Scale was designed to assess stress over time with multiple measures. In this study, participants were given the instrument only once and participants experienced the death of a parent over a broad spectrum, anywhere from 3 weeks to 2^5 years since the death. Complicated bereavement is not expressed within a certain time period, and restriction on time since death may not have provided enough flexibility to detect the effects of bereavement. It was believed that a better understanding of adaptation to loss and a more accurate assessment of stress could have been made by multiple administrations of the lES at similar lengths of time since death. By the same right, coping measures could have been tracked to determine developmental styles and utilization of new behaviors. Self -report instruments, in and of themselves, are limiting and infer bias (Lewis, 1989). There were no observational instruments used with participants, and no other instruments were used to detect expressions of

PAGE 121

Ill complicated bereavement in adolescents such as depression, academic problems, or life changes. The effect of social desirability was a concern since meetings were held with some of the participants in their homes. Participants may have felt a need to protect the family image or deny the effects of death. Implications; Counseling. Future Research, and Theory This study has implications in several areas. There are implications for counseling and for future directions for research as well as theoretical models of stress for adolescents who have experienced parental death. The average adolescent in this study reported positive levels of family functioning, yet moderate to high degrees of stress. To explore whether these finding are a function of this unique sample, the study might be replicated with groups of bereaved children in the schools. This would be a preventive measure to help children normalize their feelings of grief and to help them learn positive coping strategies. Research studies (Wass & Stillion, 1989) indicate that youth need their own support groups and that the surviving parent needs to receive help before working effectively with the family on adjustment to loss. Longitudinal studies are needed with multiple measures in order to clarify which coping resources, styles, and behaviors are important to children at different stages of development. There are few instruments that are designed to assess the normal processes of grief. Most studies have used instruments designed to measure pathological behaviors.

PAGE 122

112 many of which are considered to be normal processes of grieving. New measures are called for to help children and families understand processes of bereavement. Although the design of this research was intended to describe the adolescent experience, there is a need for further assessment of other family members through an observational measure of family behaviors that are collected by someone outside of the family. This information would lend insight about the overall adjustment of adolescents from a more systemic point of view and give a more comprehensive profile of how adolescents cope with the effects of parental death. Another recommendation for future research might be to select a sample of adolescents based on low and high levels of family cohesion as comparison groups for bereaved studies. Other samples with established polarities of adolescent or family variables might be used as comparison groups. Samples chosen by the types of death such as violent deaths as opposed to long-term illnesses might be valuable in assessing the impact of death and designing interventions for these specific groups. An assessment of life changes due to parental death is another predictor variable of stress that might be useful in treating bereaved adolescents. Selection of instruments for bereavement studies is critical to the assessment of changes due to the death. The lES was not designed specifically as a bereavement instrument and is valid in the assessment of stress for only 7 days prior to administration. Instruments designed to

PAGE 123

113 assess grief in terms of the effects of life changes might provide a more accurate profile of the bereaved adolescent. A qualitative component to research with adolescents regarding the death of a parent was important to this research. It seemed to encourage the subjects to become invested in the research, and it was designed to bring about an awareness of what they have learned through their experiences and to confirm this knowledge through the advice they would give to others. Open-ended questionnaires and personal interviews might allow the gathering of rich information and provide a way to assess the ability of adolescents to generate alternative solutions to stressful situations without having to use predetermined lists of self -report instruments. The need for alternative ways to study bereaved adolescents and families is called for, especially in light of the limited quality of instruments available for such studies. There are implications for the use of transactional models of stress theory when working with bereaved adolescents and their families. The concept of cognitive mediational processes encourages an approach which utilizes the total relationship the adolescent has with the environment to include the context in which the death occurs, the style of coping, and the availability of individual and family resources available to the adolescent (Compas, 1987). It is a respectful approach in determining the meaning of the loss for the individual. The assumption is that the individual has the ability to evaluate the personal significance of death and determine which coping

PAGE 124

114 options are available and which can be used effectively. Evaluations are assumed to be a constant interaction with the environment and personal adjustments are made through this reappraisal process. With this process in mind, coping strategies can be designed with various approaches that may appeal to different personalities of various age groups. Cognitive appraisal processes can be developed that allow the individual to learn to alter appraisals of situations or to cognitively reframe the meaning of situations, thereby rendering them harmless. In coping with the death of a parent, the relationship that the child has with the dead parent can be restructured instead of coming to an end (Silverman, 1992). The life of the parent can be celebrated and good memories preserved. The m'eaning of the loss can be reprocessed throughout the life of the child without feeling the need to be cut-off from memories of the deceased in order to "get on with life." This approach encourages family and social support, open communication about the death, and eventual integration of the death into the lives of youth. Summary The results of this study suggest that there is no relationship between the degree of stress in adolescents who have experienced parental death and the level of family cohesion and adaptability, parent-adolescent communication, and coping strategies. Subjects report that talking to friends and family is the greatest comfort and that crying and praying are also helpful.

PAGE 125

115 Although there was no statistical evidence to support relationships in the variables, the strongest predictors for adolescent adjustment to the death of a parent were the use of problem-focused behaviors, the presence of social support, and open communication with families and friends. Implications for counseling, future research, and theory were suggested.

PAGE 126

APPENDIX A LETTER TO SCHOOL COUNSELORS March 29, 1996 Dear Counselor: I have received permission from your superintendent to conduct a study with adolescents, aged 12 through 18, who have experienced a parental death over the past two and a half years. My interest is in determining resources and coping skills that have been most helpful to them. I would greatly appreciate your help in identifying these students. Presently, I am working as a youth bereavement counselor at Hospice and am a doctoral candidate at the University of Florida in the Department of Counselor Education. I have also worked as a guidance counselor in middle school and high school, and appreciate your demanding schedule. I just need the names, addresses, and phone numbers for those students you might be aware of whose parents have died over the past two and a half years. I will then make all arrangements to meet with students at home or at school. There are five questionnaires I plan to have students complete, and this will take about thirty minutes. Each participant will be receive $5. Students who are in special programs for learning disabilities and emotional disabilities will not be included in the study. I am hopeful that I will be able to Identify coping skills and resources used by these students and will be pleased to share the results with you. I will call you in the next week to answer any questions you might have regarding this project and to schedule a time when we might talk. Thank you. Sincerely, Lucia Young, EdS 116

PAGE 127

APPENDIX B PARENTAL CONSENT LETTER Dear Parent: I am writing to ask your permission for your child to participate in a research project I am conducting to find out the most important things that are helpful to adolescents while adjusting to loss. I am also enclosing an invitation to your child to participate in the study. I have been working with children and adolescents for the past three years as a counselor at the Hospice of North Central Florida, and am also a doctoral candidate in the Department of Counselor Education at the University of Florida. My research interest is in the area of parental loss and how youth adjust after loss. Four questionnaires will be administered where your child is asked to simply circle a number or enter numbers as answers to questions. There is also one short answer question to complete and a demographic questionnaire. The questionnaires are designed to identify ways of coping that have been helpful to your child. It should take about thirty minutes to complete these questionnaires, and your child will receive $5.00 for taking the time to participate. Scores will be kept confidential through a coding system, and only group scores will be available to participants. Personal identity will not be made known to anyone or appear in any writing. You or your child may withdraw from this project at any time,' and your child does not have to answer any question that he/she does not wish to answer. I will be calling you to schedule a time when I can come to your house to meet you and complete the questionnaires with your child. Alternatively, if it is more convenient for me to meet at school with your child, I will schedule a time that is acceptable to the teachers and your child. I shall be pleased to share the group results of this study with you upon completion of the project. 117

PAGE 128

118 If you would like to contact me regarding the study, please call the Hospice office at 378-3620 between the hours of 8:30 A.M. and 5:00 P.M. Please fill out the permission form below if you agree to allow your child to participate. I look forward to talking with you. Sincerely, Lucia Young, EdS I have read the procedure described above. I agree to allow my child, , to participate in Mrs. Young's study on adolescent coping, and I have received a copy of the above letter. Parent /Guardian Date

PAGE 129

APPENDIX C LETTER TO ADOLESCENT Dear I would like to invite you to participate in a project to discover how adolescents cope during stressful times. All participants have lost a parent within the past two years. Here is how it works. I will come to your home or I can meet you at your school and I will have four questionnaires for you to answer. Questions are answered by circling numbers or entering numbers as answers to the questions. There is also one question where you will be asked to give a short response in writing, and there is a short demographic questionnaire. This should only take you about thirty minutes to complete. You do not have to answer any questions that you do not wish to answer, and you will receive $5.00 for participating . You do not put your name on the questionnaires and results are confidential. I will need to code the questionnaires by entering a number on top of the papers so I can keep track of results. Your participation is, of course, voluntary and you may withdraw at any time. If you are interested in hearing about the results of this study, I shall be pleased to give you results by phone or send you a summary of results by mail. I will call you in the next week to see if you are interested and to schedule a time when we can meet. I look forward to talking with you soon. Sincerely, Lucia Young, EdS 119

PAGE 130

APPENDIX D CHILD ASSENT SCRIPT To adolescent participant: Hello, I am Mrs. Young. Did your mom/ dad/ guardian tell you that I would be meeting with you today? I would like to find out what things are helpful to youth after the death of a parent, and I have five questionnaires I would like for you to complete. It should take about 30 minutes. Is this okay with you? Do you have any questions before we begin? I'll give you all of the questionnaires and you can go wherever you would like, where you are most comfortable to work on these. Please fill them out in the order in which you receive them. You do not have to answer any questions that you do not wish to answer, and you can withdraw from the study at any time. Then, just give them to me when you are finished. (If questionnaires are completed at home) . (If questionnaires are completed at school) Here are the questionnaires I would like you to complete. Please fill them out in the order in which you receive them. Just take your time and let me know when you are finished. Here is your $5.00 for doing all of this. I would be pleased to talk with you later about my results if you are interested. Would you like me to call you and let you know? Thank you. 120

PAGE 131

APPENDIX E DEMOGRAPHIC QUESTIONNAIRE 1. Age of student 2 . Gender of student 3. Ethnicity: Caucasian African American Hispanic Asian American Other 4. County of residence: 5. Member of the family who died: Mother Father Grandmother (guardian) Grandfather (guardian) Other (guardian) 6. Age of family member who died 7. Gender of family member who died 8. What was the approximate date of the death? 9. The death was due to a long illness sudden death due to illness accident other 121

PAGE 132

APPENDIX F ADOLESCENT QUESTIONNAIRE With the death of a parent, so many changes occur. What are some of the most important things that were helpful to you during the time right after the death and things that are still helpful to you while coping with your loss? What advice would you give a friend whose parent died and had a similar experience to yours? 122

PAGE 133

APPENDIX G LETTER TO PARTICIPANTS Dear Thank you for participating in the study on adolescents who have experienced the death of a parent. Many of you asked me to contact you with the results of the study, and this is what I found. My results are based on the average scores of your questionnaires so these results may or may not apply to you. Most of you have strong coping skills and have learned through experimenting with a variety of methods. Your friends and families seem to be your most valuable support system. Results indicate that there is positive and open communication with you and your parents, and that your families are functioning well and adjusting to the loss of your loved one. The most frequent coping style chosen by you was avoiding problems" although you chose a higher number of ,1^7^^^ which tend to be more helpful in managing difficulties and stress such as talking with family members about problems and doing things together with family, following rules, praying, improving yourselves through physical activity and working hard in school, organizing yoSr life and thinking positively. ^ i ^ f^>=1^^ J^^A ^ pleasure meeting with you, and I am grateful tor all of the memories you shared with me. Sincerely, Lucia Young 123

PAGE 134

REFERENCES Adams-Greenly, M. , & Moynihan, R. T. (1983). Helping the children of terminally ill parents. American Journal of Orthopsychiatry. 53 (2). 219-229. Baker, J. E. , Sedney, M. A. , & Gross, E. (1992). Theory and reyiew: Psychological tasks for bereaved children. American Journal of Orthopsychiatry . 62.(1), 105-116. Balk, D. E. (1991). Death and adolescent bereavement: Current research and future directions. Journal of Adolescent Research. 6 (1). 7-27. Barnes, H. L. , & Olson, D. H. (1985) Parent-adolescent communication and the Circumplex Model. Child Development. 56 . 438-447. Barth, J. C. (1989). Families cope with the death of a parent: The family therapist's role. In L. CombrinckGraham (Ed.), Children in family contexts: Perspectives on treatment (pp. 299-321) . New York: The Guilford Press. Becker, W. R. (1973). The denial of death . New York: Free Press. Bertalanffy, L. von. (1975). General systems theory. In R. Brent & K. Jolan (Eds.), General systems theory and human communication (pp. 6-20) . Rochelle Park, NJ: Hayden Billings, A. G. , & Moos, R. H. (1981). The role of coping responses in attenuating the impact of stressful life events. Journal of Behavioral Medicine. 4 . 139-157. Black, D. (1991) . Family intervention with families bereaved or about to be bereaved. In D. Papadatou & C. Papadatos (Eds.), Children and death (pp. 135-143). New York: Hemisphere Publishing. Corporation. Bonnard, A. (1961) . Truancy and pilfering associated with bereavement. In S. Leonard & H. Schneer (Eds.), Adolescents: Psychoan alytic approach to problems and therapy (pp. 152-179). New York: Dell. Bowlby, J. (1980a). Loss. New York: Basic Books. 124

PAGE 135

125 Bowlby, J. (1980b) . Loss, sadness, and depression . London: Tavistock Publications. Bowlby, J. (1963) . Pathological mourning and childhood mourning. Journal of the American Psychoanalytic Association. 11 . 500-541. Brice, C. W. (1982) . Mourning throughout the life cycle. The American Journal of Psychoanalysis. 42 . 315-326. Breznitz, S., & Goldberger, L. (1991). Stress research at a crossroads. In L. Goldberger & S. Breznitz (Eds.), Handbook of stress: Theoretical and clinical aspects (pp. 3-6) . New York: The Free Press. Brown, F. H. (1988). The impact of death and serious illness on the family life cycle. In B. Carter & M. McGoldrick (Eds.), The changing family life cycle (pp. 457-482) . Needham Heights, Massachusetts: Allyn and Bacon. Buehler, C. (1990). Adjustment. In J. Touliato, B. F. Perlmutter, & M. A. Straus (Eds.), Handbook of family measure ment techniques (pp. 493-516) . London: Sage Publications. Bureau of Economic and Business Research. (1991) . Florida statistical abstr act. 1991 . Tallahassee: University Press of Florida. Cain, A. E. & Fast, I. (1966). Children's disturbed reactions to parent suicide. American Journal of Orthops ychiatry . 3 6 ^ 873-880. Cannon, W. B. (1939) . The wisdom of the body . New YorkNorton. Cheifetz, P. N. , Stavrakakis, G. & Lester, E. P. (1989) Studies of the affective state in bereaved children. Canadian Journal of Psychiatry. 34 . 688-692. Coddington, R. D. (1972). The significance of life events as etiologic factors in the disease of children — II. A study of a normal population. Journal o f Psvchosnmat i n Research . l , 205-213. Cohen, P., Dizenhuz, I., & winget, C. (1977). Family adaptation to terminal illness and death of a parent Social Casework, ^ 223-228. Coping with stress during childhood and adolescence. Psycholog ical bulletin, loi ri^. 393T v O •

PAGE 136

126 Compas, B. E. , Malcarne, V. L. , & Fondacaro, K. M. (1988). Coping with stressful events in older children and young adolescents. Journal of Consulting and Clinical Psychology. 56 (3). 405-411. Corr, C. A. (1991) . Understanding adolescents and death. In D. Papadatou & C. Papadatos (Eds.), Children and death (pp. 33-51) . New York: Hemisphere. Cragg, C. E. & Berman, H. A. (1990) . Adolescents' reactions to the death of a parent. In J. D. Morgan (Ed.), The dying and the bereaved teenager (pp. 43-49). Philadelphia, PA: The Charles Press. erase, D. R. , & Crase, D. (1989). Single-child families and death. Childhood Education. 89 . 153-156. Crook, T. , & Eliot, J. (1980). Parental death during childhood and adult depression: A critical review of the literature. Psychological Bulletin. 87 . 252-259. Crosby, J. F. , & Jose, N. L. (1983). Death: Family adjustment to loss. In C. R. Figley & H. I. McCubbin (Eds.), Stress and the family: Vol. 2. Coping with catastrophe (pp. 76-102). New York: Brunner/Mazel . Davies, B., Spinetta, J., Martinson, I., McClowry, S., & Kulenkamp, E. (1986). Manifestations of levels of functioning in grieving families. Journal of Family Issues. 7 (3) . 297-313. Dietrich, D. R. , & Shabad, P. C. (1989). The problem of loss and mourning . Madison, CT: International Universities Press. Dise-Lewis, J. E. (1988). The life events and coping inventory: An assessment of stress in children. Psychosomatic Medicine. 50 . 484-499. Ebata, A. T. , & Moos, R. H. (1991). Coping and adjustment in distressed and healthy adolescents. Journal of Applied Developmental Psychology. 12 . 33-54. Edelman, H. (1994). Motherless daughters: The legacy of loss . New York: Addison-Wesley . Elizur, E., & Kaffman, M. (1982). Children's bereavement reactions following death of the father: II. Journal of the American Ac ademy of Child Psychiatry. :>i (5) , 474izur, E., & Kaffman, M. (1983). Factors influencing the severity of childhood bereavement reactions. American Journal of Orth opsychiatry. 53 ^ 668-676.

PAGE 137

127 Elkind, D. (1968) . Egocentrism in adolescence. Child Development. 38 . 1025-1034. Erikson, E. H. (1959). Identity and the life crisis . New York: Free Press. Eth, S., & Pynoos, R. (1985). Interaction of trauma and grief in childhood. In S. Eth & R. Pynoos, R. (Eds.), Post-traumatic stress disorder in children (pp. 171186) . Washington, DC: American Psychiatric Press. Fleming, S. J., & Adolph, R. (1986). Helping bereaved adolescents: Needs and responses. In C. A. Corr & J. N. McNeil (Eds.), Adolescence and death (pp. 97-118). New York: Springer. Folkman, S. & Lazarus, R. S. (1980) . An analysis of coping in a middle-aged sample. Journal of Health and Social Behavior. 21 . 219-239. Forsythe, C. , & Compas, B. E. (1987). Interaction of cognitive appraisals of stressful events and coping: Testing the goodness of fit hypothesis. Cognitive Therapy and Research. 11 . 473-485. Freud, S. (1957a). The ego and the id . In J. Rickman, J. ( Ed . ) , A general selection from the works of Sigmund Freud (pp. 210-235) . Garden City, NY: Doubleday. Freud, S. (1957b). Mourning and melancholia. In J. Strachey (Ed. and Trans.). The standard edition of the complete psychological works of Sigmund Freud (Vol. 14, pp. 237-258) . London: Hogarth Press. (Original work published 1917) . Freudenberger, H. J., Gallagher, K.M. (1995). Emotional consequences of loss for our adolescents. Psvchothera pv 32(1), 150-153. Furman, E. (1974). A child^s parent dies . New Haven, CT: Yale University Press. Furman, E. (1983). Studies in childhood bereavement. Canadia n Journal of Psychiatry . 28 . 241-247. Furman, R. A. (1964). Death and the young child: Some preliminary considerations. Psychoanalytic Study of the Child (Vol. 19, pp. 321-333). New York: International University Press. Galvin, K. M. , & Brommel, B. J. (1982). Family communication: C ohesion and change . Dallas, TX: Scott Foresman and Company. '

PAGE 138

128 Gehring, R. M. , & Feldman, S. S. (1988). Adolescents' perceptions of family cohesion and power: A methodological study of the family system test. Journal of Adolescent Research. 3 (1). 33-52. Gergen, K. J. (1985) . The social constructionist movement in modern psychology. American Psychologist. 40 . 266275. Goodman, R. A. (1986) . Adolescent grief characteristics when a parent dies . Unpublished doctoral dissertation. University of Colorado, Boulder. Goodyear, I. M. (1990). Family relationships, life events and childhood psychopathology . Journal of Child Psychology and Psvchiatrv. 31 (1) ^ 161-192. Gordon, A. K. (1986). The tattered cloak of immortality. In C. A. Corr & J. N. McNeil (Eds.), Adolescence and death (pp. 16-31). New York: Springer. Gray, R. E. (1987) . Adolescent response to the death of a parent. Journal of Youth and Adolescence. 16 ^ 511-525. Gray, R. E. (1989). Adolescents' perceptions of social support after the death of a parent. Journal of Psychosocial Oncology. 7 (3) ^ 127-144. Greene, A. L. , & Larson, R. W. (1991). Variation in stress reactivity during adolescence. In E. M. Cummings, A. L. Greene, & K. H. Karraker (Eds.), Life-soan developmental psychology: Pers pectives on stress and cop ing (pp. 195209) . Hillsdale, New Jersey: Lawrence Erlbaum Associates. Harris, E. S. (1991). Adolescent bereavement following the death of a parent: An exploratory study. child Psychiatry and H uman Development. (4^ ^ 267-281. Hauser, S. T. , Borman, E. H., Bowlds, M. , Powers, S. I. Jacobson, A. M. , Noam, G. G., & Knoebber, K. (1991)' Understanding coping within adolescence: Ego development and coping strategies. In E. M. Cummings, A. L. Greene, & K. H. Karraker (Eds.), Life-span develoDmenta 1 psychology: Perspectives on stress and cooing (pp. 177194). Hillsdale, New Jersey: Lawrence Erlbaum Associates. Holroyd, K. A., & Lazarus, R. s. (1982). Stress, Coping and somatic adaptation. In C. Goldberger & S. Breznitz (Eds.), Handbook of stress: Theoretic al and rlinir^al aspects (pp. 21-35)^ New York: Free Press. ~

PAGE 139

129 Horowitz, M. J., Field, N. P., & Classen, C. C. (1993). Stress response syndromes and their treatment. In L. Goldgerger & S. Breznitz (Eds.), Handbook of stress: Theoretical and clinical aspects (2nd ed. , pp. 757-773). New York: The Free Press. Horowitz, M. J. Weiss, D. S., Kaltreider, N. , Krupnick, J., Marmar, C. , Wilner, N., & DeWitt, K. (1984). Reactions to the death of a parent: Results from patients and field subjects. Journal of Nervous and Mental Disease. 172(7), 383-392. Horowitz, M.J., Wilner, N., & Alvarez, W. (1979). Impact of Event Scale: A measure of subjective stress. Psychosomatic Medicine. 41 . (3), 209-218. Jackson, S., & Bosma, H. (1990a). Coping and self in adolescence. In H. Bosma & S. Jackson (Eds.), Coping and self -conce pt in adolescence , pp. 203-221. New York* Spr inger-Ver lag . Jackson, S., & Bosma, H. (1990b). Coping and self-concept; retrospect and prospect. In H. Bosma & S. Jackson (Eds.), Coping and self -concept in adolescence (pp. 203221) . New York: Springer-Verlag. Johnson, J. H. (1986). Life events as stressors in childhood and adolescence (Vol. 8). Beverly Hills California: Sage Publications. ' Kaffman, M., & Elizur, E. (1973). Children's bereavement reactions following death of the father. Journal of Family T herapy. 1 , 203-228. Kastenbaum, R. (1972). On the future of death: Some images and options. Omega . 3 , 306-318. Kastenbaum, R. j. (1986). Death, society, and huir^^r, experience. Columbus, OH: Charles E. Merill. ^^""Z' u* Bowen, M. (1988). Family Eva1n;.l-inn New York: W. W. Norton and Company. ^^^lli J^l^^^-' Parental grief: Solace and resolution. New York: Springer. ^^'-'^t'^'.Ji' Facilitation of mourning during Sjifn°°?' A. Weiner, A. H. Kutscher, D. Battin, A. Arkin, & I. K. Goldberg (Eds.), Perspectives on bereavemPnt (pp. 76-100). New Ycirk: Arn o P?el£ Konopka, G. (1980) . Coping with the stresses and strains of adolescence. Social Development Ts^n^.. , 1-17. Lazarus, R. s. (1966). Psvchol oai na l stres.-:. ;.nH ^-v,^ coping process. New York: McGraw-Hill. coping

PAGE 140

130 Lazarus, R. S. (1991a). Cognition and motivation in emotion. American Psychologist. 46 (4). 352-367. Lazarus, R. S. (1991b) . Emotion and adaptation . New York: Oxford University Press. Lazarus, R. S. (1991c). Progress on a cognitivemotivational-relational theory of emotion. American Psvcholoaist. 46 ^8). 819-834. Lazarus, R. S. (1993). Why we should think of stress as a subset of emotion. In L. Goldberger & S. Breznitz (Eds.), Handbook of stress; Theoretical and clinical aspects (2nd ed., pp. 21-39). New York: The Free Press. Lazarus, R. S., & Folkman, S. (1984). Stress, appraisal. and coping. New York: Springer Publishing Company. Lazarus, R. s., Kanner, A. D., & Folkman, S. (1980). Emotions: A cognitive-phenomenological analysis. In R. Plutchik & H. Kellerman (Eds.), Emotion: Theory. research, and experience: Volume 1. Theories of emotion (pp. 189-217). New York: Academic Press. Lazarus, R. s., & Launier, R. (1978). Stress-related transactions between person and environment. In L. A. Pervin & M. Lewis (Eds.), Perspectives in interactional psychology (pp. 287-327). New York: Plenum Press. Lewis, J. M. (1989). The birth of the familv: An empirical mguiry. New York: Brunner/Mazel . Lifschitz, M., Berman, D., & Galili, A. (1977). Bereaved children. Journal of American Academy of Child Psychiatry. lf> (7 \ ^ 272-284. Lukas, C, & Seiden, H. (1987). Silent grief: Living in the wake of suicide . New York: Scribner. Maddi, S. (1981). Individual development: Its significance for stress responsitivity and stress adaptation. Personality development. In C. Moore (Ed.), Adolescence and stress. (DHHS Publication No. ADM 81-10098 pp III7II3). Washington, DC: U.S. Government Printing' Office. ^ Mccubbin, H., & Patterson, J. (1983). The family stress process: The double ABCX model of adjustment and adaptation In H. I. McCubbin, M. B. Susman, & J. m. S ocial stress and thP family: Adv;,nn^c. and development in familv stress t heory and rT^^i;:;:^; (PP7-37). New York: Haworth. McGoldrick, M. (1991). Echoes from the past: Helping families mourn their losses. In F. Walsh & M. McGoldrick (Eds.), Living beyo nd loss: n^ath in th^ family (pp. 50-78). New York: W. W. Norton. ^^^"^

PAGE 141

131 McGoldrick, M. , & Walsh, F. (1991). A time to mourn: Death and the family life cycle. In F. Walsh & M. McGoldrick (Eds.), Living beyond loss; Death in the family (pp. 3049) . New York: W.W. Norton. McNeil, J. N. (1986). Talking about death: Adolescents, parents, and peers. In Corr, C. A. & McNeil, J. N. (Eds.), Adolescence and death (pp. 185-201). New York: Springer Publishing Company. McNeil, J. N. (1989). Death education in the home: Parents talk with their children. In H. Wass & C. A. Corr (Eds.), Childhood and death (pp. 293-313). Washington, DC: Hemisphere. Menaghan, E. (1983) . Individual coping efforts and family studies: Conceptual and methodological studies. In H. McCubbin, M. Sussman, & J. Patterson (Eds.), Social Stress and the Familv (pp. 113-135). New York: Haworth. Miller, J. B. (1971). Children's reactions to the death of a parent. Journal of the American Psvchoanalvtic Association. 19 . 697-719. Moody, R. A., & Moody, C. P. (1991). A family perspective: Helping children acknowledge and express grief following the death of a parent. Death Studies. 15 . 587-602. Moos, R. H., & Billings, A. C. (1982). Conceptualizing and measuring coping resources and processes. In L. Goldberger & S. Breznitz (Eds.), Handbook of stress (pp. 212-230). New York: Free PresT. Moos, R. H., & Schaefer, J. A. (1991). Coping resources and processes: Current concepts and measures. In L Goldberger & S. Breznitz, S. (Eds.) Handbook of stress: Theoretical and clinical aspects (2nd ed. , pp. 234-257) New York : The Free Press . rr y • Moriarty, A., & Toussieng, P. (1976). Adolescent cnninrr . New York: Grune and Stratton. — ^ i'.^i(1^9°)The dying a nd the bereaved teen;.rT^-r . Philadelphia: The Charles Press. johsx. Nagera, H. (1970). Children's reactions to the death of important objects: A developmental approach. Psychoanalytic Study of the chiiri (pp. 360-400) . New York: International Universities Press. Noller, P & callan, J. (1986). Adolescent and parent perceptions of family cohesion and adaptability Journal of Adolescence, 97-106.

PAGE 142

132 Noller, P., Seth-Smith, M. , Bouma, R. , & Schweitzer, R. (1992) . Parent and adolescent perceptions of family functioning: A comparison of clinic and non-clinic families. Journal of Adolescence. 15 . 101-114. Olbrich, E. (1990). Coping and Devlopment. In H. Bosma & S. Jackson (Eds.), Coping and self-concept in adolescence (pp. 35-47) . New York: Springer-Verlag. Olson, D. H. (1989) . Circumplex model of family systems VIII: Family assessment and intervention. In D. H. Olson, C. S. Russell, & D. H. Sprenkle (Eds.), Circumplex model; Systemic assessment and treatment of families (pp. 7-49) . New York: Haworth. Olson, D. H., & McCubbin, H. I. (1983). Families: What makes them work. Beverly Hills, CA. : Sage Publications. Olson, D. H., McCubbin, H. I., Barnes, H. , Larsen, A., Muxen, M., & Wilson, M. (1992). Familv inventories . St. Paul: Family Social Science. Olson, D. H., Portner, J., & Bell, R. Q. (1982). FACES II: Familv adaptabi lity and cohesion evaluation scales . St. Paul, MN: University of Minnesota, Family Social Science. Olson, D. H., & Stewart, K. L. (1991). Family systems and health behaviors. In H. E. Schroeder (Ed.), New directions in health psy chology assessment (pp. 27-64). New York: Hemisphere. Osterweis, M. Solomon, F. & Green, M. (Eds.). (1984). Bereavement; React ions, conseauences . and care . Washington, DC: National Academy Press. Parkes, C. M. , & Weiss, R. S. (1983). Recovery from bereavement. New York: Basic Books. Patterson, J. M. (1988). Families experiencing stress: I. The family adjustment and adapatation response modelII. Applying the FAAR model to health-related issues for intervention and research. Family Systems MedininA , Patterson, J. m. , & McCubbin, H. I. (1987). Adolescent coping style and behaviors; Conceptualization and measurement. Journal of Adolescence, in , 163-186. Patterson, J. M., & McCubbin, H. I. (1991). A-COPE: Adolescent coping orientation for problem experiences. In H. I. McCubbin & A. I. Thompson (Eds.), Famil y assessment inventories for rps^ arch and p r-^;:FT7:^p^^ eg., pp. 235-250). Madison; University of Wisconsin

PAGE 143

133 Pearlin, L. , & Schooler, C. (1978). The structure of coping. Journal of Health and Social Behavior. 19 . 221. Piaget, J. (1959). The language and thought of the child . (M. Gabain, Trans.). London: Routledge and Kegan Paul, Ltd. Public Health Service, Centers for Disease Control, National Center for Health Statistics. (1991) . Vital statistics of the U. S. . 1988; Vol. 2. Mortality. Part A . Hyattsville, MD: U.S. Department of Health and Human Services. Raphael, B. (1982). The young child and the death of a parent. In C. M. Parkes & J. Stevenson-Hinde (Eds.), The Place of at tachment in human behavior (pp. 131-150) London: Tavistock. Raphael, B. (1983). The anatomy of bereavement (pp. 147176) . New York: Basic Books, Inc. Rizzuto, A. M. (1979). The birth of the living God: A psychoanalytic study. Chicago: University of Chicago Rosenblatt, p., & Elde, C. (1990). Shared reminiscence about a deceased parent: Implications for grief ?nA-?^?;°" counselling. Family Relations. 39 . ^-S?^!!' (^986). Approach, avoidance, and coping with stress. American Psvcholng i si-, ai , 813-819. Rubin, s. S. (1985). The resolution of bereavement: a clinical focus on the relationshp to the deceased. l!!''!!?! Theor y, Res e arch . T raining ;,nH p>-.^^^^^ ""'"ISigSsi^: ^^il^-n Of -ck parents. Maudsley ^^^tl'ri i'' ^ Skolnick, N. (1992). Childhood parental death and depression m adulthood: Roles of surviving parent and family environment. American .Tnnm.i jj^ "^ ^^""^"^ Orthopsychiatry, fi? , 504-516^ "

PAGE 144

134 Sandler, I. N., West, S. G., Baca, L. , Pillow, D. R. , Gersten, J. C, Rogosch, F. , Virdin, L. , Beals, J., Reynolds, K. D., Kallgren, C. , Tein, J., Kriege, G. , Cole,E., & Ramirez, R. (1992). Linking empirically based theory and evaluation: The family bereavement program . American Journal of Community Psycholocry. 20(4), 491-521. Segal, R. M. (1984). Helping children express grief through symbolic communication. Social Case Work. 26 . 590-599. Selye, A. (1993). History of the stress concept. In L. Goldberger & S. Breznitz (Eds.), Handbook of stress; Theoretical and clinical aspects (2nd ed. , pp. 7-17)). New York: The Free Press. Selye, H. (1956). The stress of life . New York: McGrawHill. Shapiro, E. R. (1994). Grief as a family process: A develop mental approach to clinical practice . New York: The Guilford Press. Shuchter, S. R. (1986) . Dimensions of arief: Adjusting to the death of a spouse . San Francisco: Jossey-Bass. Siegel, K. , & Gorey, E. (1994). Childhood bereavement due to parental death from acguired immunodeficiency syndrome. Developm ental and Behavioral Pediatri cs. 15(3), 66-S48! Siegel, K, Mesagno, F. P., & Christ, G. (1990). A prevention program for bereaved children. American Journal of Orthopsychiatry. 60 (2), 168-175. Silverman, P. R. , Nickman, S., & Worden, W. (1992) Detachment revisited: The child's reconstruction of a dead parent. American Journal of Orthons vnhiai-T-^/ 62 (4), 494-503^ ~ ^ Silverman, P. R. , & Worden, J. w. (1992). Children's reactions m the early months after the death of a 104^'^^* Journal of Orthopsychiatry. 62 m, 93Silverman, S. M., & Silverman, P. R. (1979). Parent-child communication m widowed families. American Journal of Psychothe rapy. 22 C\ \ ^ 428-441. Spirito A., Stark, L. J. , & Williams, C. (1988). Development of a brief coping checklist for use with 4r''555^574 ^^^^"""^ • Journal of Pediatric Psy^h^i^^y,

PAGE 145

135 Spivack, G., & Shure, M. B. (1982). The cognition of social adjustment: Interpersonal cognitive problem-solving thinking. In B. B. Lahey & A. E. Kazdin (Eds.), Advances in clinical child psychology (Vol. 5, pp. 323372) . New York: Plenum Press. Spivack, G., & Shure, M. B. (1985). ICPS and beyond: Centripetal and centrifugal forces. American Journal of Community Psychology. 13 . 226-243. Stern, M. , Van Slyck, M. R. , & Newland, L. M. (1992). Adolescent development and family dynamics: Delineating a knowledge base for family mediation. Mediation Quarterly. 9 (4^. 307-322. Swearingen, E. M. , & Cohen, L. H. (1985). Life events and psychological distress: A prospective study of young adolescents. Developmental Psychology. 21 (6) ^ 10451054 . Tyszkowa, M. (1991). Coping with difficult school situations and stress resistance. In H. Bosma & S. Jackson (Eds.), Coping and self-concept in adolescence (pp. 188-201). New York: Springer-Verlag. U.S. Bureau of the Census. (1990). Statistical abstract nf the U.S. : 199? (112th ed.). Washington, DC: U.S. Department of Commerce. Valliant, G. (1977). Adaptation to life . Boston: Little Brown . Van Eerdewegh, M. M. , Bieri, M. , Parilla, R. , & Clayton, P. (1982). The bereaved child. British Journal of Psychiatry. 14 0 . 23-29. Van Eerdewegh, M. M. , Clayton, P. J. , & Van Eerdewegh, P. (1985). The bereaved child: Variables influencing early psychopathology. British Journal of Psychiatry. 147 . 188—194 . Vida, S., & Grizenko, N. (1989). DSM-III-R and the phenomenology of childhood bereavement: A review. Canadian Journal of Psychiatry. 34 ^ 148-155. ^°^Now' yl'v^.\^l^^^^ ; Linking objects and l.nVi ng Phennn.^n. , New York: International Universities Press. ^°^^??^;^^•',T,S^"''^''^' Pi^^l^er, L., & Williams, W. (1971). The reactions of family system to sudden and unexpected death. Omega. 2 rn/ini-inA ^^^Jm,;nii J^^^-^^' Reconceptualizing family stress. Journal of Marnaap and thP Family. 47 , 827-837.

PAGE 146

136 Wass, H. , & Stillion, J. M. (1988). Death in the lives of children and adolescents. In H. Wass, F. N. Berardo, & R. A. Neimeyer (Eds.), Dying: Facing the facts (pp. 201-208) . Washington, DC: Hemisphere. Werner, E. , & Smith, R. (1982). Vulnerable but invincible: A longitudinal study of resilient children and youth. New York: McGraw-Hill. Wills, T. A. (1986) . Stress and coping in early adolescence: Relationship to substance use in urban school sample. Health Psychology. 5 . 503-529. Wolfenstein, M. (1966) . How is mourning possible? study of the Child. 21 . 93-123. Youniss, J., & Smollar, S. (1985). Adolescent relations with mothers, fathers and friends . Chicago: University of Chicago Press. Zilberg, N. J., Weiss, D. S., & Horowitz, M. J. (1982). Impact of Event Scale: A cross-validation study and some empirical evidence supporting a conceptual model of stress response syndromes. Journal of Consulting and Clinical Psychology. 50 (3^ 407-414.

PAGE 147

BIOGRAPHICAL SKETCH Lucia Patat Young was born in Charleston, South Carolina. She received a Bachelor of Science degree in psychology from the University of South Carolina in 1969 and a Master of Education and Specialist in Education degree from the University of Florida in 1991. After she graduated from the University of South Carolina, she became a mental health associate in the adolescent psychiatric unit at the Medical University of South Carolina. She moved to Boston, Massachusetts, and was the administrative assistant to the chairman of biochemistry at Harvard College. Her love of science led to a teaching position in the biological sciences at Brimmer and May School, a private girls' school in Chestnut Hill. After her two children, David and Allison, were born, she was involved in volunteer work for various charitable organizations. She became an editorial assistant to the editor of an international scientific journal. She then entered graduate school in 1989. She has worked as a guidance counselor and as a mental health counselor in the fields of addictions, sexual abuse, and eating disorders with children, adolescents, and adults. She joined the staff at Hospice as a children's bereavement counselor and is currently a licensed mental health counselor. She has a private practice in Gainesville, Florida. 137

PAGE 148

I certify that I have read this study and that in my opinion it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degree of Doctor of Philosophy. Ellen S. Amatea, Chair Professor of Counselor Education I certify that I have read this study and that in my opinion it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degree of Doctor of Philosophy. M. David Miller Associate Professor of Foundations of Education I certify that I have read this study and that in my opinion it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degree of Doctor of Philosophy. Jam^s/H. Pitts Asspciate Professor of Counselor Education I certify that I have read this study and that in my opinion it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degree of Doctop-ef.^ijLpsophy . Robert Profes This dissertation was submitted to the graduate Faculty of the College of Education and to the Graduate School and was accepted as partial fulfillment of the requirements for the degree of Doctor of Philosophy. August 1996 Dean, Collegjg of Educati Dean, Graduate School