J

.. .-
":33'3‘:

 

‘ri‘

“ “flaws

"-1 .M- H};

“ ‘4‘!-
«H \;
-::u..“4 ‘3'";{7‘
"m" {Ca ".3:
. “I”

 

. A14

’ I.
41'2” ;. W1,

um.“
. Mun.-

' N'-
:~,~‘.L 5‘9?

 

'"p."
' I“.
‘.
“-2.. g.

7...“:

 

 

3:: run?

‘_ {3‘ ‘Qt' x...:£*“
.;

 

Yfiv;€;

.
m,-

$3
‘-‘..“ ”111»
4;. ~

um z».

'1 J- 3'1»

q...
1:" ‘ 19w."-

e
”‘4

1: . . 1‘ m:3.vu;£“ Ni 3

k
"r?

 

3-0 «3"?

 

" flint.

' '1
. -‘ . . .357. ' :3“
7‘1

 

ufiQL‘I'NV

“I”...

u

 

I 'a {‘4

5
k
:1
E
1

 

V t
. I .1. "13.1”!"
'1 I ma, 0'

‘ 4:075.
f" I
fin}...

I

:4; .
a .' . :..,

. 29'
'{w

 

:w'.
,M':;; '
11’ {a '3‘ '1‘

"ADC!"
cS-‘g v
era." ‘

k. ‘15.
haw-7E

.

J‘ “in:

In. 5}.-
‘H" v:

. .R

m-...-..

x-..“ -

5..

flaw! h.

.,..
u-

 

I~~~.%

~7Mu.o.w....
‘9

“:3;-

      

‘54 l‘l -
W? 'mkxcg‘mrr.
u... _~

‘ ‘14.”:
51- ~45
-t~ ft.“

 

::'€ L-W’Lhfia“ ‘5
o £1“.th
‘1'. V
U

u.‘
’1

‘ V
'17,,

w <(,

. ~ _
-..I, :1?»

L...
~ a.»

(1M -3? ....
«L.—
w‘l' A.‘?.].‘f~. ““71:

h
I
“9?: .

“mm 1‘
, M

.‘ '
,~. *
MIRR.
no.
.~

3.--...“

     

 

 

“7* :fflt‘
“D“; 1;; flit-335k
‘KWLW_K

,~~~i

’lhl‘ -JLL
- L .

v.

 

. I 4;.

H' "W.

 

3" ,Y,‘;.' ‘

I'IIA';-I

 

MICHIGAN STATE UNIVERSITY LIBRARIES

This is to certify that the

dissertation entitled

HIGH SCHOOL SENIORS PERCEPTION OF
GRADUATION: A STUDY OF SEPARATION

presented by

MARGARET ROSE CONNOLLY

has been accepted towards fulfillment
of the requirements for

Ph. D. degree in Psychology

 

4 ////f m
/

Majorropfessor ”W
John P. McKinney

March 16. 1992
Date

 

MS U i: an Affirmative Action/Equal Opportunity Institution

 

 

 

 

 

 

 

LIBRARY
Michigan State
University

 

 

 

PLACE IN RETURN BOX to remove this checkout from your record.
TO AVOID FINES return on or before due due.

DATE DUE DATE DUE DATE DUE

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

MSU is An Affinnetive Action/Equal Opportunity institution
chMt

 

HIGH SCHOOL SENIORS PERCEPTION OF GRADUATION:
A STUDY OF SEPARATION

By

Margaret Rose Connolly

A DISSERTATION

Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of

DOCTOR OF PHILOSOPHY

Department of Psychology

1992

J i
and adui
attachmc
"Cgative
rB-surfac
SCparatjc
childhoo

In
qUestiOm
hYPOthes
Depressii
control g
Supponin

ABSTRACT

HIGH SCHOOL SENIORS PERCEPTION OF GRADUATION:
A STUDY OF SEPARATION

By

Margaret Rose Connolly

John Bowlby (1969) extended his theory of attachment into adolescence
and adulthood suggesting that early separations affect the formation of later
attachments as well as one’s reactions to later separations. He suggested that
negative affect associated with early separations from attachment figures could
re-surface at later separation experiences. High schOol graduation often is a
separation from family and friends as well as from a school system and from
childhood.

In the current study high school juniors and seniors were given
questionnaires two to three weeks prior to graduation. Contrary to the
hypothesis, the seniors’ scores on the Center for Epidemiologic Studies
Depression inventory (CBS-D) were not as high as were the scores of the
control group of juniors. Females’ scores were higher than males’ scores.
Supporting the hypotheses were significant correlations between lower CES-D

scores (for seniors) and positive opinion of self, higher family relatedness, and

ii

 

 

higher re
between
As predi
very diff
CES-D s
of self c
subject,
than 5%
Correlat:
expressi
away fr
necessa
CXplana
COnside
PFeVioL
separat

resDon.

higher relatedness to peers. Failing to support the hypothesis was a correlation
between lower CES-D scores and a larger number of separation experiences.
As predicted seniors’ expectations that the year following graduation would be
very different from the senior year was significantly correlated with higher
CBS-D scores. A step-wise, multiple regression analysis indicates that opinion
of self contributes the most to the variance in CES-D scores with sex of
subject, family relatedness and number of previous separations contributing less
than 5 % each. The expectation that the next year would be very different
correlated significantly with plans to live away from home. Also, subjects
expressing relatively more positive self opinions tended to report plans to live
away from home the next year. It appears that previous separations do not
necessarily have a negative impact on current separations. A possible
explanation for the more positive effect of previous separations that takes into
consideration the self esteem factor also, is that successfully dealing with
previous separations provides one with a sense of competence for handling new
separation situations. It is suggested that several factors interact to affect one’s

response to separation.

iii

 

 

LIST (

INTRC

METH
RESU]

DISCL

APPEI

APPE}

APPEI

LIST (

TABLE OF CONTENTS

LIST OF TABLES .................................... v
INTRODUCTION .................................... 1
Literature review ................................. 6
Attachment theory ............................ 6

Related research ............................ l9

Hypotheses .................................... 35
METHOD ........................................ 39
RESULTS ........................................ 50
DISCUSSION ...................................... 67
Comclusions ................................... 84
APPENDIX A: Juniors’ Questionnaire Instructions ............. 89
Seniors’ Questionnaire Instructions ............. 90

General Information Questionnaire ............. 91

Graduation and Future Plans Questionnaire ........ 93

Friends Questionnaire ..................... 94

APPENDIX B: Letter to Juniors and Their Parents ............. 95
Letter to Seniors and Their Parents ............. 97

APPENDIX C: Instructions to Public School Seniors ............ 99
LIST OF REFERENCES .............................. lOO

iv

LIST OF TABLES

Correlation Matrix ............................... 52

Results of Step-wise Multiple Regression
Analysis for Seniors .............................. 54

Seniors’ Responses to Separation History Items ............. 56

Means, Standard Deviations and t-values for
Seniors’ Relatedness Variables ........................ 58

Means of the Depression Scores (CES-D)

by Sex and Year ................. , ............... 6 0
Percent of Responses to Selected Graduation

and Future Plans Questions .......................... 61
Projected Living Arrangements ....................... 64
Distance Seniors Plan to Live from Home Next Year .......... 65
Plans for the Next Year ............................ 66

 

 

 

Tl
attachme
The pun
reactions
prelude '
friends,

J<
1973, 1;
Children
Children
and Wa
resDons.
resPOns

SChOO] i

INTRODUCTION

Throughout life we face many separations that change or disrupt
attachments we have made. High school graduation is one of these separations.
The purpose of this research is to identify important variables in adolescents’
reactions to high school graduation. High school graduation often is the
prelude to leaving home and family as well as school, classmates, teachers and
friends.

John Bowlby proposed a theory of attachment and separation (1969,
1973, 1980, 1982, 1988). Other psychologists have studied the responses of
children of different ages to separation. It is the separation experiences of
children under school age that have been studied in depth. Ainsworth, Blehar,
and Waters (1978) and Mahler, Pine, and Bergman (1975) studied children’s
responses to separation from parents. Field (1984) investigated preschoolers’
responses to separation from peers. Fewer studies have examined responses of
school age children to separation. College students’ perceptions of leaving
home have been studied by several individuals (Moore and Hotch, 1981, 1983;

Moore, 1987; Sullivan and Sullivan, 1980). But, other than reactions to death

1

 

 

 

and dim
attentim
m

(
and 19S
individt
attachm
Specific
669). ;
individi
individl
from p2
Parents‘
forms I:
behaVio
from Iii
Child”S .
%

A

and fOr

lmpma]

2

and divorce, high school students’ responses to separations have received little
attention.
W

On the basis of the writing of John Bowlby (1969, 1973, 1980, 1982,
and 1988) it is apparent that the kinds of relationships that exist between
individuals affect the response to separation. Bowlby (1982b) defines
attachment as a strong disposition "to seek proximity to and contact with [a
specific] individual and to do so especially in certain specified conditions" (p.
669). Attachment relationships are ongoing relationships in which one
individual perceives another as stronger and wiser and seeks closeness to that
individual in certain situations. Infants and toddlers often protest separating
from parents, especially from mother. Toddlers begin to anticipate their
parents’ leaving and protest before the parents leave. Protest can take many
forms but all are attachment behaviors. Bowlby (1982b) defined attachment
behavior as ”...any of the various forms of behavior that the person engages in
from time to time to obtain or maintain a desired proximity" (p. 670). The
child’s protests appear to be an attempt to control the parent’s behavior.
WM

Attachment relationships are important to humankind both for individual
and for species survival. The adult-infant bond has long been recognized as

important for infant survival among all species of mammals, yet even as an

3

infant the individual begins to separate from the parent or primary caregiver.
The response to separation is different if the child is the initiator, is in control,
and moves away from the parent to explore. Thus there are two categories of
separation. Bowlby (1988), Bretherton (1985), and Main, Kaplan, and Cassidy
(1985), among others, believe that attachment theory can explain both the
tendency to seek proximity to a specific individual when feeling threatened and
to explore when the environment is non-threatening.

As the child grows, attachments begin to be directed toward persons
outside the family and to groups and institutions such as church or school
(Bowlby, 1969). Bowlby suggests that the attachments that develop during
adolescence and adulthood should be seen as a natural continuation of the
attachment behavior of childhood. In adolescence the individual develops
attachments to friends and perhaps to school. Therefore, we would expect to
see proximity seeking behaviors as responses to the separation or the anticipated
separation signaled by graduation. We can expect to see responses associated
with loss also. With the ability that develops for abstract thinking and
anticipating future events there is reason to expect responses in anticipation of
graduation. Again because of expanded cognitive capacities, we can expect that

the responses to separation and loss may take various forms.

W

According to attachment theory a variety of factors affect one’s ability to
deal with separations (and loss). One of these is one’s history of separations.
Perhaps the most influential experiences are the separation experiences that
occur between six months and three years. However, later separation
experiences can be important also. Among later separation experiences there
are some that are predictable such as starting school, the ending of a school
year, going to camp, moving, and graduating. Some separation experiences are
not predictable, for example, the death of a close family member, parent
divorce, or the death of a close friend or long-time classmate. Some
separations are relatively brief and expected; when starting school a child must
separate from the parent/caregiver. Experience shows that brief and expected
separations, such as starting school, are seldom accompanied by strong
emotion. On the other hand, unexpected or permanent separations such as
death of a family member often result in anxiety, depressed mood, and sadness.
Relatively prolonged separations such as may occur with parental divorce may
cause sadness, depressed mood, or anxiety also. High school graduation is a
relatively permanent albeit expected separation for most adolescents.

Other important factors in dealing with separation include one’s
perception of resources, both personal and interpersonal, self-esteem and the

sex of the individual. If one has dealt with other crises successfully and has a

5

strategy for doing so, the separation may not be as traumatic as if one has no
strategy. If one has experienced appropriate support and assistance when
needed there is probably an expectation that resources will be available again
when needed. Adolescent and adult women seem more concerned with the
establishment and maintenance of interpersonal relationships than are men
(Douvan and Adelson, 1966; Miller, 1976; Gilligan, 1982); for women the
breaking of important relationship bonds is related to depressive disorders
(Rutter, 1986b). Self-esteem, as reflected in opinions about one’s self, has
been identified as an important factor in studies dealing with separation and
studies of depressive behavior (Moore, 1987; Kandell and Davis, 1982;
Allgood-Merten, Lewinsohn and Hops, 1990). It seems reasonable that one’s
expectations about life following a separation would influence one’s reactions
too.

The current study is designed to identify the variables related to high
school seniors’ reactions to graduation. Teachers and parents of high school
seniors frequently refer to ”senioritis”. When asked to specify behaviors
related to senioritis in an informal survey, six local teachers mentioned inability
to concentrate, to finish assignments, or to maintain the same quality of work;
reports of feeling depressed, tired and sad; being concerned about not being
able to see their friends everyday; increased impulsivness or irritability. I am

suggesting that these changes are a reaction to the anticipated separations that

6

will accompany and follow graduation and that the emotions underlying these
reactions are anxiety, sadness, and depressed mood.

I am going to study this issue by comparing high school juniors’ and
seniors’ responses on the Center for Epidemiological Studies-Depression
Inventory (CES—D) and by doing a step-wise multiple regression analysis to
determine the relative importance of 1) attachment to parents, 2) previous
separation experiences, 3) attachment to friends, 4) extent of expected change,
5) self esteem, and 6) sex. The literature on separation suggests that these
factors are important in one’s reaction to separation.

Literature Review
Attachment Theory

John Bowlby became interested in children’s reactions to separation from
parents in 1927 when he worked in a home for emotionally disturbed boys.
After becoming a child psychiatrist and psychoanalyst he established a special
research unit at the Tavistock Clinic in London to study children’s responses to
separation from their mothers. The Study of separation led to the study of the
development of attachment. He proposed a theory of attachment in 1969;
subsequent works (1973, 1980, 1988) have dealt with separation and loss.
Others (Mahler et al., 1975; Sroufe and Waters, 1977; Ainsworth et al., 1978;
Bretherton, 1985; and Main et al., 1985) have studied the nature of attachment

and the factors involved in its development and maintenance.

7

According to Bowlby (1969), attachment is based on an internal control
system characterized by setting, set-goal, and feedback. Set-goal is a goal
determined by the kind of environment in which the animal has been operating
during its evolution. In mammals some of these systems have unvarying set-
goals; blood sugar and temperature regulation are two examples of such
systems. The set-goal of temperature regulation for most humans is 98.6
degrees Fahrenheit. Bowlby suggests that a type of control system called a
servo-mechanism is applicable to attachment. Servo-mechanisms involve
repeatedly or continuously changing set goals to which the system must adjust.
The ability to kick a moving soccer ball or hit a moving baseball or the ability
of a child to keep a moving parent in view is taken as an indication that this
kind of system occurs in humans.

According to Bowlby, proximity is the set-goal of the attachment system.
The biological function of proximity is protection from predators (harm). The
young child seeks closeness to the attachment figure when alarmed. Although
proximity seeking is activated by isolation or alarm, activation is not automatic
and does not invariably occur when a given distance from the attachment figure
is surpassed or a stranger arrives. For example, a strange child may be greeted
with curiosity by a toddler whereas a strange adult elicits anxiety and proximity

seeking.

8
Stimulated by Bowlby’s work, Sroufe and Waters (1977) suggested that

the goal of the system may be better identified as "felt security" (p. 1186). As
the term implies, felt security is a subjective experience based on individual
evaluation of the situation; it must be inferred from observable behavior.
Labeling the set goal ‘felt security’ rather than proximity makes it clearer that a
variety of internal and external variables are involved. (Important parameters
include setting, familiarity, preceding events, individual’s mood and
developmental level.)

Attachment behaviors are those that facilitate proximity to the attachment
figure but they are not unique to the attachment situation. According to
Bowlby attachment behaviors form a class or category. Sroufe and Waters
(1977) remind us that behaviors within a class or category can substitute for
each other at different times and at different developmental levels. Thus it is
that the crying, crawling, clinging behaviors of the infant later give way to
speech, letters and telephone calls. The latter may temporarily replace face to
face interaction for adults.

In modern western society where individuals frequently move away from
home, devising means of maintaining closeness to attachment figures becomes
important and has become a big business. The economic status of companies
that enable people to communicate over distances indicates that humans place

importance on maintaining contact with attachment figures. The mobility of

9

individuals in our world makes us aware, also, of the value of forming new
attachments. The ability to maintain existing attachments and to form new ones
makes moving to a new community less frightening.

There are strong parallels between the attachment behaviors of children
and those of adults. Adults, too, may be demanding of others and in conditions
of danger or disaster seek the proximity of a special person. A secure partner
relationship can provide security for personal risk-taking or exploration in
adulthood just as it does in childhood.

When a child becomes capable of cognitive perspective taking he or she
is able to understand something of parent motives and plans and is able to
manipulate parents to bring their plans into closer agreement with his/her own.
This is an example of what is meant by a goal correction mechanism.

Attachment is directed toward specific individuals. Individual
recognition (discrimination) and highly differentiated behavior are involved. As
Lorenz dramatically demonstrated, this is as true in the parent-young relations
of birds and mammals as it is in humans. Ainsworth (1989) describes the
attachment figure as an individual who "is important as a unique individual and
is interchangeable with none other" (p. 711).

The growth of attachment to a specific individual during the first 3-12
months is well documented; subsequent development is less well documented,

yet evidence indicates that it persists until puberty in many species and past

lO
puberty in some (Douvan and Adelson, 1966; B108, 1967; Bowlby, 1969; Main

et al., 1985). Between infancy and adulthood attachments develop to family
members other than the primary caregiver, to regular baby-sitters and to peers
(Ainsworth, 1989; Bowlby, 1969; Weiss, 1982; Marris, 1982; Freud and Dann,
1951 [cited in Bowlby, 1982a] and Field, 1984). During early school years
many children grasp the parent’s hand while out and usually seek the parent, or
substitute, if something goes wrong while playing or if frightened. According
to Bowlby (1969) adolescents’ attachment to parents changes as attachments
form to peers and other adults. (Ainsworth, 1989, suggests this phenomenon
may be mediated by "hormonal, neurophysiological and cognitive changes" p.
710.) Bowlby (1969) suggests that adolescents and adults may develop
attachments to groups and institutions as well as people. (Such attachments
may be mediated by a person within the group.) In old age attachment
behavior, out of necessity, may be directed toward younger persons.

Bowlby ( 1969) refers to the common observation that preschool children
continue to show intense attachment behavior through their second and third
years. During the eleventh or twelfth month the child begins to anticipate
mother’s departure and to protest anticipated separation. According to Bowlby,
the conditions necessary for a child of three or older to tolerate separation from
mother include availability of familiar, subordinate attachment figures, good

health, knowledge of mother’s whereabouts, confidence that mother will return

11

soon and a feeling of safety (i.e., the child is not alarmed). The increasing
ability to tolerate separation from mother has been documented by Murphy et
a1. (1962) and others.

Between their third and fourth birthdays most children become capable of
understanding something of the parent’s motives and plans and of manipulating
parents’ plans to bring them into closer agreement with their own. "Conceptual
perspective taking" (Bowlby, 1982a, p. 368) is responsible for the ability to
manipulate parents. White (1966) suggests that success in manipulating the
environment, including people, leads to a feeling of competence.

Cognitive abilities play an important role in developing attachment
relationships. Initially there is discrimination of the attachment figure; this is
followed by evidence of object permanence as the child attempts to re-establish
contact with the absent attachment figure or as he anticipates the mother
leaving. Bowlby (1973) employs the concept internal "working model," (p.
203) or mental representation, to help explain discrimination, separation
anxiety, seeking the absent mother and manipulating parents to achieve his own
goals. Bretherton (1985) conceives of a working model as a mental
representation of an aspect of the environment, the self, others, or relationships
to others that are important to the individual. The data for the model come
from the individual’s experiences and once the data are incorporated into the

model they are used to evaluate experiences and guide behavior. Internal

12

working models can be reorganized to take account of new and even
contradictory experiences. The child compares his current perception of a
person or situation with his internal working model. A discrepancy between
the model and the perception may cause the child distress. The model is
modified and the feelings of distress are included in the revised model along
with other information about the situation or person.

Internal working models of the self-in-attachment-relationships are
formed from experiences in attachment relevant events, especially those in
which there is an intention to seek caregiver proximity. The internal working
models contain both affective and cognitive components (Bretherton, 1985).
Once formed the internal working models tend to be stable, operate
automatically, and exist outside consciousness. The self-in-separation models
continue to be influenced, however, by experience and by information received
from parents (caregivers) and from others about them. Internal working models
influence later relationships and experiences.

In infancy and childhood attempts to gain proximity that are consistently
accepted lead to a different internal working model than do attempts which are
consistently blocked or are accepted unpredictably. Children who see
themselves as loveable and valuable have grown up in stable homes. They
have been encouraged to become autonomous. There is frank communication

from parents whose working models of themselves, the child, and others are

l3

valid and are open to question and revision. These children are able to form
attachment bonds with people outside of their own families (Bowlby, 1973).
Bowlby says it is important for an individual, of any age, to feel "confident that
an attachment figure will be available to him whenever he desires it" (p. 202).
This feeling of confidence or trust makes the person less prone to fear or
anxiety than he/she would be without it.

Very young children do not have the capacity for planning and reflection
that adolescents have. However, the mental representation of an attachment
figure may help children cope with separation. An internal representation
provides a sense of security and enables the child to venture further and further
from the secure base provided by the attachment figure (Sroufe and Waters,
1977 ; Bowlby, 1988). Although the period between six months and five years
is said to be the most sensitive for development of expectations of attachment
figure availability, the expectations, or mental representations, continue to be
modified by experiences throughout adolescence.

Although internal working models continue to influence relationships and
experiences, adolescents’ ability for formal thought enables them to think about
their own thinking, to step outside a relationship and see it operating. This is
very important because it makes it possible for adolescents to understand the
influence of past events on current relationships and to evaluate their responses

within relationships. They are able to report on their own attachment

14

relationships. Adolescents are capable, also, of planning ways to bridge
distances to maintain attachment relationships.

Bowlby (1973) places individual differences in adolescent and adult
attachment on a continuum. At the extremes are those who cut themselves
completely off from family and those who are so strongly attached that they
cannot or will not direct their attachment to others. Bowlby (1969) suggests
that most adolescents fall in the middle with enduring family bonds and strong
attachment bonds to others also.

In summary, attachments occur first to one’s primary caregiver. This is
usually the mother. Some time later attachments develop to others, usually the
father, siblings, others who live in the home or people with whom the child
associates frequently. Internal working models are formed of these
relationships. These models change as the individual has new experiences with
attachment figures. The models influence later relationships. The most
enduring attachments are the attachments to the mother and father.

The internal working models of family relationships contain an
expectation of security and support or the expectation of insecurity. The
expectation is built on the individual’s personal history of separation and
reunion experiences. All internal working models include the affective aspects
of separation and reunion experiences, namely the grief and anxiety associated

with separation as well as the joy of reunion. They include, also, a sense of

15

security which allows the child to venture further and further from the parent.
As the child grows and spends more and more time with peers and adults
outside the home, attachments may develop to friends, teachers, or the school.
The attachment will be indicated by the individual’s desire to spend time with
these people or at the school.

Since family attachments are usually the most enduring, it is expected
that they will have the most influence on seniors’ reaction to graduation and
those seniors who have positive relatedness to their parents will have lower
scores on the Center for Epidemiological Studies-Depression inventory (CES-
D). However, an individual who is strongly attached to parents but has no peer
attachments may not feel secure enough to venture beyond the family. Bowlby
(1969) suggests that some individuals have had traumatic or numerous
separation experiences that make their development of attachment relationships
difficult. Graduation would, therefore, pose a greater threat. For the
individuals who have had numerous separation experiences the CES-D score is
expected to be higher than the CES-D scores of individuals with fewer
separation experiences.
52mm

Bowlby (1973), Mahler et a1. (1975), and Bretherton (1985) suggest that
early separation experiences affect later experiences through the internal

working models. In the discussion of attachment, it was suggested that early

 

 

16

experiences can provide an internal model that includes feelings of security and
allows the growing child to move into the world and form attachments to
others. Bretherton (1985) suggests that the affective component of the
separation and loss model includes sadness and anxiety. This has been
supported by the work of Ainsworth et al. (1978), Mahler et a1. (1975) and was
discussed by Bowlby (1969, 1973). However, separations that result in
satisfactory adjustment and positive experiences become part of the mental
representation also.

Main et a1. (1985) expanded Bowlby’s theory by proposing a theory of
separation based on mental representation of the attachment relationship and
suggesting that reunion responses following short and long-term separations
could be seen as indicative of the child’s view or "internal working model of
the relationship" (p. 68). They suggest that working models are organized
schematically by action and action outcomes (including attempts to regain
parent in parent’s absence); hence, each person’s internal working model of an
attachment relationship is unique. Each person’s working model of separation
is unique also. However, the essential differences between representations of
attachment and separation are presumed to be limited in number.

Internal working models influence behavior and are influenced by events

throughout life. Experiences such as a change of nursery school (Fields, 1984)

17

or transition from junior high to high school (Blyth, Simmons and Carlton-
Ford, 1983) influence the models.

Separations that occur during childhood usually lead to what
psychoanalytically oriented psychologists call individuation or a sense of self as
an individual, separate from those around him/her. A number of these
psychologists (e.g., Blos, 1967; Josselson, 1980; Laufer, 1966; Stierlin, 1974)
suggest that adolescence is a time of further individuation. There are two
significant contributions to this new sense of self. First is an increased ability
to physically separate, for example the ability to drive a car enables the
adolescent to go further and more often without parent accompaniment.

Second, the cognitive ability of formal or abstract thought allows adolescents to
think about their own thinking and to think about themselves. As individuation
and autonomy develop, so does a sense of identity; a part of that sense of
identity is what one thinks about one’s self.

Wm, Adolescents are usually seen as moving away
from the family. However, for an individual of any age to feel confident
enough to move away from current attachments he/she must have a sense of
separateness from and relatedness to the world, i.e., a sense of self. Marcia
(1980) referred to it as an awareness of one’s "own uniqueness and similarity to

others" (p. 159).

18
Mahler et a1. (1975) referred to the development of individual identity

as individuation and suggested that later separations (such as those experienced
in adolescence) may re—activate unresolved individuation issues and anxiety
through a mechanism similar to Bowlby’s internal working model. Laufer
(1966), Blos (1967), Stierlin (1974), and Josselson (1980) agree that work on
individuation is resumed in adolescence. Josselson states that "there is an
interdependent sequence of: individuationlautonomy/identity formation" (p.192).
Identity formation involves integrating the various "aspects of the self that have
become individuated and autonomous " (p. 192) into a coherent whole. When
"good enough individuation" (p.191) is achieved there is the feeling of making
one’s own choices and being responsible for what happens.

Laufer (1966) and Blos (1967) discuss the importance of adolescents
separating (detaching or disengaging) from internalized objects (parents) in
order to form extrafamilial attachments. Stierlin (1974) describes separation in
adolescence as "a transactional process between. . .parents and children" (p. 3).
Ideally this process is a "gradually expanding spiral of mutual individuation and
differentiation occurring on various emotional, cognitive, and moral levels" (p.
3).

Although individuation and identity formation are important aspects of
adolescent development and may play a role in ability to separate from family,

they are not the focus of this study. However, further research on the

l9

interaction between level of individuation and feelings about graduation could
prove interesting.
R l h

Interestingly, most of the work on attachment has focused on behaviors that
occur while the individual is separated from the attachment figure or upon
reunion after separation. This has been true of research on adolescence too.
This includes the various studies of home leaving done by Moore and Hotch
(1981, 1983), Moore, (1987) and Sullivan and Sullivan (1980) and the research
on adolescents’ responses to parent divorce or to death (Moore and Hotch,
1982; Proulx and Koulack, 1987; Irion, Coon, and Blanchard-Fields, 1988;
Wallerstein and Kelly, 1980). Little research has dealt with anticipated
separation.

Bowlby (1973) suggests that early separation experiences affect later
experiences. Empirical support is found in a study by Stein, Levy, and
Glasberg (1974) who found significantly more early childhood and antecedent
separations among both black and white, male and female suicide attempters
than in control groups. Early childhood was defined as between birth and age 7
years; antecedent separations were disruptions of close interpersonal
relationships within one year of the attempted suicide. Barnes and Prosen
(1985) reported that adults whose fathers had died in the 0 to 6 and 10 to 15

age ranges had the highest depression scores on the CES-D (Center for

20
Epidemiological Studies Depression Scale.) Bloom (1977/ 1978) found that past

separation experiences, the quality of parent-child relationships and the ability
to care for one’s self affected the adolescent-parent separation process. Podell
(1989) found that some adolescent survivors of peer death were struggling with
reconciliation of earlier losses, the memory of which was reactivated by the
traumatic deaths of their peers.

Family stability, A study by Hansburg (1972) suggests that all children
have histories that can result in concern over separation. Hansburg (1972) gave
children and young adolescents (ages 10 to 15 years) the Separation Anxiety
Test. He found that those who had grown up in stable families responded to
pictures of separation with distress and concern two to three times more often
than with anger and blame. Children from rejecting families gave nearly equal
distress/concern and anger/blame responses. To pictures representing a major
disruption, children from disrupted homes gave responses indicating that the
child would do his best to get along or that he’d be happier following the major
disruption. Bowlby (1973) suggests that the latter responses indicate inadequate
attachments and that the underlying unconscious models may create difficulty in
later relationships.

Moore and Hotch (1982) investigated the impact of divorce and ordinal
position on college students’ perceptions of having left home, feelings about

leaving home, conflict associated with home leaving, the importance of personal

21

control factors and emotional separation. Parental divorce was associated with
emotional separation factors for males. Studies by Kelly and Wallerstein
(1976) and Wallerstein and Kelly (1980) found that boys of all ages experience
more adjustment problems after parental divorce than do girls. The results of
McCombs and Forehand (1989) support Wallerstein and Kelly’s finding (1980)
that amount of conflict between ex-spouses and mother-adolescent conflict are
significant factors in adolescents’ adjustment following divorce. Although the
results were not statistically significant there was a tendency for first born
females to report that feelings of having left home were related to feelings of
personal control (Moore and Hotch, 1982).

Proulx and Koulack (1987) found that college students from separated
families did not experience more conflict when leaving home than did students
from intact families. Furthermore the type and amount of divorce-related
conflict was not related to higher emotional separation or locus of control.
When divorce-related conflict was openly expressed, feelings of personal
control increased and feelings about leaving home became more positive.
Again, emotional separation was significantly higher for males and for students
from separated families.

Irion, Coon and Blanchard-Fields (1988) compared adolescents 14-17
years old from divorced and intact families and concluded that "adolescents of

divorced parents were more vulnerable and appraised situations less favorably

22

the more stressful they were" (p. 143). The adolescents from intact families
reacted to greater stress by using more active, adaptive strategies, planned
problem solving and positive reappraisal. Children from intact families tended
to see stressful situations as an opportunity. In challenging situations the intact
family group felt more in control. This suggests that children of divorced
families might react with more negative feelings to a later separation experience
such as graduation.

Adoleseeot’s reaction to death of others, There is relatively little written
that specifically speaks to adolescents’ reactions to the death of others.
However, Garmezy (1986) reports that loss events such as parental death or
divorce are marked by grief and bereavement although reactions vary in
severity from individual to individual.

For adolescent survivors of another’s suicide the questions of guilt over
not being able to prevent the death and concern regarding one’s own potential
for suicide are very potent and add to the adolescent’s feelings of vulnerability
and grief (V alente and Sellers, 1986). Podell (1989) refers to "survivor guilt"
(p.71). Survivor guilt is characterized by feelings of excessive responsibility
and magical thinking. The magical thinking may involve "regression to an
earlier developmental state of organization around prior childhood losses of
significant objects" (p. 72). These include losses other than those of family

members.

23
Wolfelt (1987, 1988), himself a survivor of adolescent peer suicide,

stresses the importance of reconciliation rather than resolution in dealing with
death. He suggests that this is a need for adolescents just as it is for adults.
He makes the point that one never "gets over" a death and that sadness lingers
with the memory.

Much of the literature that reports adolescents’ reactions to another’s
death seems to focus on symptoms of depression or identification of
(pre)suicidal behaviors. For example, Grollman (1987) cites the following
behaviors and feelings as relevant: loneliness, isolation, worthlessness,
hopelessness, depression, prolonged boredom, loss of appetite, insomnia or
excessive sleeping, physical complaints, difficulty concentrating and angry
outbursts. Fleming and Adolph ( 1986) reported the following cognitive
responses to the death of a loved one: "I am marked"; "1 am different . . . .I
am bereaved" (p.105). "1 can’t handle this" (p. 106). "I am alone. . .no one
can really understand me"; "I can’t risk/trust because I could lose the ’other’"
(p. 108). Behavioral responses include acting out, aggression, exaggerated
independence, over investment in another or withdrawal from others. Affective
responses include depression with either flat affect or agitation, anger and
resentment, feeling unlovable. Although the authors suggest that adolescents

may experience conflicts about asking for help, most adolescents talked to their

24

mothers or family about the death. Those who talked to someone had fewer
problems following the death than those who had no one with whom to talk.

Adolescents understand the biology of death, i.e., it is final, universal,
irreversible, and involves the cessation of biological functioning (Melear,
1973). There is a simultaneous awareness that a relationship has been
disrupted. Adolescents go through the same grieving process as adults, if
allowed to do so.

Mobility, Moving often results in the disruption of attachments to
friends and home. Grief and symptoms of depressed mood have been
associated with geographic mobility for adolescents and adults. Gibbs (1985)
reported a significant relationship between geographic mobility and depression
in urban adolescent females. Depression was significantly related to parental
occupation, geographic mobility and number of self reported problems. Over
60% of the girls lived in single parent families, usually with the mother. Gibbs
cited other studies, e.g., Shaw (1979), Tooley (1970), and Werkman (1979)
indicating that frequent moves were disruptive for adolescents.

Examining the issue of mobility from a different perspective, Fried
(1963) studied adults who had moved from their homes in Boston’s West End
because of urban renewal. Both women and men experienced grief, some for
as long as two years. Grief was associated with being separated from their

friends and relatives. Those who experienced the most grief were those whose

25

closest friends had lived in the same neighborhood whereas those whose closest
friends had lived elsewhere were less likely to be sad or depressed. Fried
suggested that the grieving associated with geographic mobility is related to the
loss of "a sense of spatial identity (which) is fundamental to human functioning"
and to disruption of a ‘group identity’ (p.156). Group identity refers to a sense
of belonging to or being a part of "larger human or social entities" (p.157).
SehooL Csikszentmihalyi and Larson (1984) reported that adolescents
spend thirty-two percent of their waking hours in school. (They spend forty-
one percent of their time at home with at least thirteen percent of that time
spent alone.) Those who are active in clubs and organizations may spend even
more time at school. Those who participate in sports, band, etc. are often
strongly identified by others as being part of the school. They are often
referred to by the name of the school mascot, e.g., Huskies, Tigers, or
Spartans. Barker and Gump (1964) found that students in small schools were
more likely to be active in a wide range of activities. Students in small schools
were more likely than those in large schools to report doing things that help
them develop skills and the ability to work with others. Everyone in the small
schools seemed to feel a sense of involvement and obligation. Perhaps this
sense of involvement leads to a feeling of attachment. It may be that those

students who are very active in their school, whether large or small, develop an

26

attachment to the school. If so, they may respond to graduation with more
negative feelings than those who have not identified strongly with the school.

In summary, grief and behaviors suggestive of depressed mood have
been described as accompanying divorce, death, and moving or geographic
relocation. I am suggesting that high school graduation is a separation
experience that may re-activate the memories of previous separations and may
account for the sadness that seniors experience along with the excitement and
joy of graduation. Since nearly everyone has had some separation experiences
and seniors have the cognitive ability to anticipate the loss, I expect high school
seniors to report more depressive symptoms (have higher scores on the Center
for'Epidemiological Studies Depression Scale [CBS-DD than juniors who are
not facing the separations. In addition, I expect that those seniors who have
experienced more losses (of attachments) to have higher CES-D scores than
those whose separations have been relatively brief, temporary, and have not led
to loss. I expect that the number of separations will contribute less to the
variance of the CES-D scores than will family relatedness.

It may be the case that, as Bowlby has suggested, students may have an
attachment to their school. This could make up a small portion of the variance
on the CES-D also.

MW After reviewing the

research on depression in adolescence we can make several generalizations.

27

First, adolescents do experience depression and often evidence the same
symptoms as adults (Strober, 1983; Rutter, 1986). Second, Puig-Antich
(1986), Carlson and Cantwell (1980) among others, have indicated that children
over the age of 8 can give reliable and valid reports of their depressive
symptoms. Although Rutter (1986b) and Angold, Weissman, John et al. (1987)
found that parents and teachers failed to notice depression in young people who
reported feeling severely depressed during an interview, Angold et al. reported
that parents were sensitive to suicidal ideation. Third, since most symptoms of
depression are internal and seldom discussed, adolescents and their families
often fail to detect them (Emery, Bedrosian and Garber 1983). A questionnaire
with statements indicative of depression or depressed mood or an interview
asking similar questions may be necessary to elicit the admission of negative
feelings.

Kovacs and Beck (1977) differentiate depressed mood which is a
temporary, dysphoric mood state and depression which is a clinical syndrome
of longer duration. They suggest that psychosomatic and behavioral complaints
are age appropriate and culturally accepted expressions of depressive symptoms
for adolescents. Thus a questionnaire that includes questions about changes in
eating and sleeping habits, crying spells, etc., follows this line of thinking.

The CES-D includes such questions.

28

In terms of Bowlby’s theory of attachment, the number of stressful
separations experienced (for example, death of attachment figures, abandonment
by an attachment figure as a result of divorce, or loss of friends because of a
move) is related to the feelings of grief or depressive responses to later
separations by re-activation of the internal separation model.

Although some studies do not report significant sex differences in
depression, Allgood-Merten, Lewinsohn and Hops (1990), Roberts, Andrews,
Lewinsohn and Hops (1990), and Roberts, Lewinsohn and. Seeley (1991) report
that adolescent females score higher on the CES-D than do adolescent males.

Attaehment to frienoe, Csikszentmihalyi and Larson (1984) reported that
one fifth of an adolescent’s waking hours is spent with the family but only a
portion of that is with parents. Fifty-two percent of their time is spent with
peers. Later research (Larson and Richards, 1991) supports the earlier work of
Montemayor and Brownlee (1987), i.e., adolescents spend increasingly less
time with the family from elementary school through high school. Larson and
Richards compared pre-adolescents with older adolescents. The older subjects
spend half the amount of time with family that younger subjects spend. Older
adolescent boys spend more time at home alone whereas older adolescent girls
divide the extra time between time alone and more time with fiiends. Older
subjects were more positive about the time spent with friends than were

younger subjects.

29

Douvan and Adelson (1966) suggested that adolescent girls were more
concerned with establishment and maintenance of interpersonal relationships
whereas boys were more concerned with occupation and ideology. Miller
(1976) says that "women stay with, build on, and develop in a context of
attachment and affiliation with others. . ., for many women the threat of
disruption of an affiliation is perceived not just as a loss of a relationship but. .
.a loss of self" (p. 83). Youniss and Smoller (1985) reported 66% or more of
the friendships of their female subjects were characterized by intimacy and trust
while less than 45 % of the male friendships were so characterized. Rutter
(1986b) reported that there is evidence indicating that stress events such as
those involved in the breaking of important relationships are related to
depressive disorders (p.512). However, Rutter also suggests that anxieties
about the future may increase during adolescence; this may have implications
for males similar to the disruption of relationships for females. Although
Greenberg, Siegel, and Leitch (1983) found attachment to parents to be greater
than to peers it was suggested that it does not mean that peer relationships are
unimportant. Youniss and Smoller (1985) reported that adolescents felt closer
to their friends than their parents and 70% of them believed their friends
understood them better than did their parents. It is unlikely that many
adolescents would give up their friends without protest, suggesting that

adolescent peer relationships are meaningful. However, the research of

30
Greenberg et al. (1983) and Youniss and Smoller (1985) suggest that

adolescents are aware of the interest and concern of their parents.

W A number of the studies cited have identified sex
differences in the importance of relationships and in response to separations.
On the basis of Bowlby’s model, it is not expected that the frequency or
intensity of depressive symptoms (CES-D score) will be significantly related to
sex of subject. The experiences in the self-in-separation model are individual
specific but not related to sex of the individual. However, there are suggestions
in the research literature that attachment relationships, including those to
friends, are more important for females than for males and that disruption of
these relationships is taken more seriously by females. Therefore, I am
suggesting that high school senior females will have higher CES-D scores than
high school senior males.

Opinion of self, The opinion one has about one’s self or self-esteem,
has been identified as a contributing variable in various studies of separation
and relatedness to family and friends. Moore (1987) found that greater self-
esteem was related to the self-governance factor and a feeling of having
achieved separation from family. Self-govemance refers to making one’s own
decisions without having to have those decisions affirmed by parents. For
females, disengagement was important for feeling independent of (separated

from) family and was associated with greater self-esteem. Disengagement

31

refers to being physically away from parents and being free of parental
direction. Greenberg et al. ( 1983) looked at the association of self-esteem and
affective quality of relationships with parents and peers among adolescents 12
to 19-years-old. They found that, on measures of self-esteem, quality of
attachment to parents seemed to be a moderating factor in periods of high
stress. Graduation could be considered a period of high stress because of the
many changes associated with it. (Stress can occur with positive as well as
negative events, Selye, 1974). Opinions about the self are measured in this
study with the Rosenberg Self Esteem Scale (Rosenberg, 1965). It is expected
that the Opinion of Self, or the RSE score, will contribute significantly to the
CES-D score and that the RSE score will be positively related to the parent
relatedness index.

College students and "leaving home". Moore and Hotch (1981) studied
separation in terms of feelings about leaving home. They had college freshmen
rate 22 adjectives "selected as being a fairly comprehensive sampling of
important components of home-leaving" (p.4). A hierarchical cluster analysis
resulted in the following clusters: personal control (make own decisions),
economic independence, residence (live in a different place), physical
separation, school affiliation (consider school to be home), dissociation (won’t

go back home in the summer), emotional separation (don’t "belong" at home

32

anymore), and graduation. College freshman perceived personal control and
economic independence as the most significant markers of home leaving.

Moore and Hotch ( 1983) had college students rate a random order of 20
definitions of home leaving in terms of personal importance. Personal control
was again seen as very important; emotional separation and dissociation were
seen as unimportant; the other factors were viewed neutrally.

Moore (1987) reports two further studies on these same issues. College
freshman and sophomores completed a questionnaire on which they rated 34
statements on a seven point scale reflecting how important the issues within the
statements were to them in deciding if they had left home or separated from
parents. An oblique factor analysis found the following factors: self
governance, emotional detachment, financial independence, separate residence,
disengagement, school affiliation, starting a family and graduation.

Self governance was rated as most important, with females rating it more
important than males. Graduation, starting a family and financial independence
received high ratings also. Emotional detachment was rated as least important.
Disengagement which was described as having some psychological and physical
distance from parents was rated as moderately important. Sullivan and Sullivan
(1980) compared male college students living at home to those living away at
school. Those who go away to school exhibit "increased affection,

communication, satisfaction, and independence in relation to their parents" (p.

33

93). Some high school students may have similar attitudes toward high school
graduation, financial independence, and starting a family.

The second part of Moore’s project was concerned with whether or not
the students felt they had achieved separation. College freshman and
sophomores rated the separation factors identified in study 1 in terms of the
personal importance and attainment. Moore indicated that " correlates of
achieving separation according to subjective definition are likely to depend on
the extent to which the component is important to the individual" (p. 300).
High rankings of school affiliation were associated with greater life satisfaction
and less loneliness. When financial independence was rated as important, self-
esteem was high.

Some sex differences were found. Importance and attainment of
detachment were associated with perceived parental abandonment and feelings
of loneliness, low self-esteem, low identity achievement and difficulty leaving
home for males. High ratings of disengagement were related to perceiving
their mothers as feeling abandoned and to loneliness in males. For females,
however, high ratings of importance and attainment of disengagement were
associated with increased self-esteem, life satisfaction, ego identity achievement
and little loneliness.

As we grow we experience many separations; some separations are

temporary and we return to the attachment figures, others result in change of

34

attachment figures or the context of the attachment relationship. Bowlby
(1969). Ainsworth et al., (1978), Mahler et al., (1975) among others suggest
that the child moves further from the home base at his or her own pace.
Graduation may mean moving away from the "home base" but often the
distance is determined by the adolescent. A greater anticipated distance may be
seen as a greater separation. I am suggesting that those seniors who are
planning to move away from their hometown, thus anticipating more breaks in
attachment relationships, will have higher CES-D scores than those who are
moving away from home but not out of town. These individuals will have
higher CES-D scores than those who plan to continue to live at home.
However, it is expected that extent of anticipated change will contribute less to
the overall variance of the CES- than will relatedness to parents, friends, or
number of previous separations.

In summary, the purpose of this research is to identify important
variables in adolescents’ reactions to high school graduation. The proposed
study is designed to identify the nature of a high school phenomenon frequently
referred to by teachers and parents as senioritis. I am suggesting that the
changes in behavior are a reaction to the anticipated separations that will
accompany and follow graduation and that the emotions underlying these
reactions are anxiety, sadness, and depressed mood. The factors that will be

investigated are relatedness to parents, number of previous separations from

35

attachment figures, relatedness to friends, self-esteem, sex of subject, extent of
expected change, anticipated distance from home, and relatedness to school.
Hypotheses

Separation often results in sadness and anxiety (Bowlby, 1980; Mahler et
al., 1975; Hansburg, 1972, 1980). The anticipation of separation can result in
depressive symptoms or depressed mood also (Field, 1984). Graduation
signifies a separation with a way of life that students have been experiencing for
twelve years. Most of them will have spent much of that time with the same
peers and will have made many friends (Csikszentmihalyi and Larson, 1984).
In addition, as high school students many will have developed an identification
with the school. Graduation also signals a soon to be realized
independence/separation from family as many of the seniors will be leaving
home. Even though most seniors will have talked about looking forward to this
independence, it is a separation and it is expected that they will experience
sadness and depressed mood as graduation approaches.

Hypothesis 1: High school seniors will have higher scores on a

depression index known as the Center for Epidemiological Studies
Depression inventory (CES-D) than will high school juniors.

Previous research (Bowlby, 1980; Stein et al. 1974; Ainsworth et al.

1978; Mahler et al. 1975; Greenberg et al. 1983) suggests that early separation

36

and attachment experiences are important factors in determining one’s reaction
to current separations. Some common separations include moving, preschool,
parental separation or divorce, traveling without family; a less common but
relevant experience would be the death of a family member or close fiiend.
According to Bowlby (1969), a number of separations such as these may make

the individual more vulnerable to experiencing symptoms of depression.

Hypothesis 2: The number of separations a high school senior has had
(Biographical Information) will be positively related to his/her
CES-D score.

Hypothesis 3: Seniors’ CES-D scores will be positively related to having
experienced the death of a family member or close friend prior to
age seven (General Information question 8).

Friedrich, Reams, and Jacob (1981) found the highest incidence of
depression in the 18 to 24-year-old group. This age group is the one that
usually deals with the separations and life style changes that follow graduation.
The seniors in this study who expect "their lives to be quite different in the next
year from the past several years are more likely to report symptoms of
depressed mood than are the seniors expecting little or no change.

Hypothesis 4: High school seniors’ scores on the CES-D will be

positively related to the anticipation of substantive changes in their
lives in the coming year (Future Plans question 7).

37

High school seniors who see themselves as making their own choices
seem to have strong relationships with their parents (Hotch, 1980) and a
stronger sense of who they are (Josselson, 1980). Self-esteem has been related
to the ability to handle stress situations. Graduation is a major turning point in
the lives of high school seniors but it is one they have been anticipating and
preparing for. Those students whose personal history includes strong family
relatedness should show relatively fewer symptoms of depressed mood.

Hypothesis 5: High school seniors’ parent relatedness scores (Family

Relatedness Index) will be negatively related to their CES-D
scores.

Adolescent females tend to score higher than adolescent males on
depression measures (Allgood-Merten et al., 1990; Roberts et al., 1990;

Roberts etal., 1991).

Hypothesis 6: Female high school seniors will have higher CES-D
scores than male high school seniors.

Students often identify with their school because of their involvement in
its activities and because the community identifies them with the school. Group
identity is considered a part of adolescent identity. Bowlby (1969) suggests that

attachments can form toward groups and institutions such as schools.

38

Graduation, in a sense, means relinquishing that identity without providing a
ready substitute. Even those students going on to college are not yet identified
with that college to the extent that they have been with their high school. Fried
(1963) found that individuals who were identified with their neighborhood and
whose closest friends lived in that neighborhood experienced the most grief
when forced to relocate due to urban renewal. Graduation occurs whether or
not one is ready for it and it often means leaving familiar territory and friends
for unfamiliar territory and new friends.

Hypothesis 7: Seniors’ CES-D scores will be positively related to their

report that they will miss their high school (Graduation and Future
Plans question 11).

METHOD

Subjects

The subjects were juniors and seniors from a public and a private high
school in southern Indiana. There were 180 seniors in the private school and
360 seniors in the public school. The entire questionnaire was completed by
281 seniors; 84 were from the private school and 197 were from the public
school. The sample represents 47% of the seniors in the private school and
55% of the seniors in the public school. There were 148 juniors in the private
school and 374 juniors in the public school. The 55 juniors from the private
school completed the entire questionnaire and provided gender information; 236
juniors from the public school completed the questionnaires but data on gender
were not available. The sample represents 37% of the juniors in the private
school and 63% of all the juniors in the public school.

More females than males volunteered to participate in this study. There
were 168 senior females and 113 senior males. There were 22 junior males
and 33 junior females for whom sex of subject information was obtained.

These subjects were from the private school. The 3:2 ratio obtained for the

39

40

seniors is consistent with other studies using high school volunteer subjects
(Fischer, 1981; Yanchyshyn and Robbins, 1983). (The male-female ratio in the
schools from which the sample for this study was drawn was almost 1:1 for
each school.)

The students in both schools are primarily white. The public school
student body is 97.5% white, 1.0% black, and 1.0% Hispanic, Asian/Pacific
Islander, and 0.5% American Indian, Eskimo, or Aleut. The most recent data
tabulated indicates that the private school student body is 93.5% white, 2.5%
black, 3.0% Hispanic, and 1.0% Asian/Pacific Islander.

Families of the private school students were described by school officials
as being predominantly middle and upper middle class. It is primarily a
commuter school and draws many students from three counties as well as a
couple of students from each of three or four more nearby counties. Families
of the public school students were predominantly middle class also. According
to the US. Bureau of Census median income in the county in which these
students live was $21,379 (1988 figures). Eighty-two percent of the homes
were owner occupied. Another indication that the majority of students in the
sample are from the middle class is that three-fourths of them are going on to
school or college.

The scale for measuring depressed mood (CES—D) is influenced by

current situations. To rule out a general influence, other than graduation, the

41

school counselors were contacted. The counselors indicated that nothing
unusual had occurred in the schools during the two weeks preceding or the
week of testing. However, it is unknown if anything unusual had occurred at
home or within relationships.

Materials

The senior’s questionnaire is made up of the following parts: General
Information, Graduation and Future Plans, Opinion of Self, the Center for
Epidemiological Studies Depression Scale (CBS—D), Family Relatedness Index,
and a Relatedness to Friends Index. Choice of questions was influenced by
John Bowlby’s work on attachment, separation, and loss (1969, 1973, 1980)
and the work of Hansburg (1972 and 1980). The juniors’ questionnaire
includes the CES-D and the Relatedness to Friends Index. Copies of the
instructions for the juniors and seniors, the General Information questions, the
Graduation and Future Plans questions and the Friends questions are provided
in Appendix A.

The Center for Epidemiological Studies Depression Scale (CES-D) was
used as the measure of depressed mood. The CES-D was "designed to measure
the current level of depressive symptomatology, with emphasis on the affective
component, depressed mood" (Radloff, 1977, p.385). It is used to gather
information about feelings, thoughts and behavior "during the past week."

Current situations influence individual responses. Because this study is

42

concerned with depressed mood rather than the clinical syndrome of depression,
the CES-D is considered an appropriate measure. Moreover, choice of the
CES-D was based, in part, on the sensitivity of local school officials to the
mention of the word ’suicide’ or a reference to killing oneself in materials
given to students.

The CES-D is suitable for group use, takes approximately 5 minutes to
administer and is hand scored (Sweetland and Keyser, 1966). The scale is
considered suitable for use with African-Americans and whites as well as with
males and females over a wide age range (Radloff, 1977). Weissman,
Sholomskas, Pottenger, Pursoff and Locke (1977) found no significant
correlations of CBS-D score with age, social class or sex except for
schizophrenic patients of higher social class and female drug addicts who had
significantly higher CES-D scores (p< .05) than the average individual.

Roberts, Andrews, Lewinsohn and Hops (1990) reported internal
consistencies of .87 to .92 in a sample of over 2000 ninth through twelfth-grade
students. T est-retest correlations ranged from .49 to .60 on follow up ranging
from one week to one month. Radloff (1977) compared the CES-D with other
scales designed to measure symptoms of depression, i.e., the Lubin, the
Bradbum Negative Affect and the Bradbum Balance. Discriminant validity
ranged from .43 to .63 with these scales. Roberts, Lewinsohn and Seeley

(1991) report a CES-D sensitivity of 85.70 and a specificity of 69.40. That is,

43

the CBS-D will detect a number of false positives and may miss screening out
some true negatives. It more accurately detects current episodes of depression
than past episodes.

Roberts, Lewinsohn and Seeley (1991), Allgood-Merten, Lewinsohn and
Hops (1990), Roberts, Andrews, Lewinsohn, and Hops (1990) and Doerfler,
Felner, Rowlison, Evans and Raley (1988) investigated the use of the CES-D
with high school students. High school students have higher CES-D means
than do adults. In the studies cited 12% to 50% or more of the subjects scored
one standard deviation or more above the adult mean on the CES-D. Achieving
a score one standard deviation above the mean involves reporting many
symptoms of depression occurring at a lower frequency or several symptoms
occurring almost daily. Roberts et al. (1991) evaluated the sensitivity and
specificity of the CES-D using a receiver operating characteristics (ROC)
analysis. A CES-D score of 24 or more was optimal for determining clinical
depression in adolescent populations.

The Rosenberg Self-Esteem scale (1965) is a 10-item Guttman scale
designed to measure adolescents’ self acceptance. The Guttman scale
reproducibility coefficient is .92. Test-retest reliability has been reported at
r=.85 and r=.88 (Rosenberg, 1979, p.292). Construct validity was based on
the significant relationship between scores on the self-esteem scale and scores

on a depressive affect measure of anxiety developed by the Research Branch of

44
the US. Army in World War II, and by peer group reputation. Crandall (cited

in Rosenberg, 1979) reported that the RSE correlated .60 with the Coopersmith
Self-esteem Inventory. The ten items are answered on a four point scale from

strongly agree to strongly disagree. The items are scored only as agreement or
disagreement with agreement indicating a lower opinion of self.

The Family Relatedness Index, based on the work of Murphey, Silber,
Coelho, Hamburg and Greenburg (1963) has a reliability of .90 (Hotch, 1980).
The Family Relatedness Index was used to measure seniors’ feelings about their
families.

The lO-item Relatedness to Friends index was developed for this study
and is based on Hotch, 1980. Split half reliability computed on the current data
was .82.

General Information question number 8 provides the separation history
index. The separation history index is a sample of separation experiences and
is not a comprehensive list. Item content was based on the writing of Bowlby
(1969, 1973) and Hansburg (1972, 1980). Experiences ranged from having
"vacationed" out of one’s home state, to death of an immediate family member.
(See Appendix B.) The list of experiences was sent to eight child psychologists
who teach and/or have clinical practices in Southern Indiana. Five of the eight
replied; all who replied said the list appeared to be a good sample of childhood

separation experiences. Separations occasioned by accident or illness were

45

suggested; loss of appearance or limbs as the result of accidents or illnesses
were suggested also. Because these are rare (as indicated by those who
mentioned them) the list was not amended to include them. The separation
score is the sum of the items checked by the respondent to question number 8.
If there is more than one death in a category, each death is counted. For this
study the mean number of separations was 4.75 (s.d. =1.53). The number of
years since the death was ignored in determining the score. Number of
separations due to death was entered into the analysis separately as well as
being included in the total separation score.

The Graduation and Future Plans section and the General Information
section of the questionnaire were developed using Hotch’s questionnaire as a
model. Individual questions of interest for this study make up the items.
Included are questions based on the work of Bowlby (1969), Fried (1963) and
Sullivan and Sullivan (1980).

Questions 5, 6, and 7 from the General Information section of the
questionnaire were included for information about plans for the coming year
and were worded as follows:

5. Next year I plan to (t) live with my family, (2) have my own

place, (3) live in another town.

6. If you will be away from home, how far away will you be? (1) up

to 50 miles, (2) 50-100 miles, (3) 100-200 miles, (4) over 200
miles.

46

7. Next year I plan to be (l_) working full time and supporting
myself, (2) in the military service, (3) going to college or
technical school, 1‘1) other.

Numbers in parenthesis indicated coding for statistical analyses. In addition, a
score of O was assigned to question 5 for those few who indicated some other
option. For those living at home, item 6 was coded 0 (no data).

Questions 7 and 11 of the graduation and Future Plans questionnaire
were of interest for hypotheses 4 and 7 respectively. Question 7 is the
respondent’s subjective expectation about his/her life after graduation (next
year). Item 11 reflects the respondent’s attachment to school. The questions
were worded as follows:

7. I think next year my life will be very different from this year.

Strongly agree/Agree/Disagree/Strongly disagree.

11. I think I am going to miss this school very much. Strongly
agree/Agree/Disagree/Strongly disagree.

Responses to items 7 and 11 were coded as follows: Strongly agree=4,
Agree=3, Disagree=2, Strongly disagree=1.
fitment:

All students in the junior and senior classes were given the opportunity

to respond to the questionnaire. Those who did not wish to participate were not

47

given the packet. All responses were completely anonymous; students wrote
neither their names nor any student numbers on the questionnaire booklets. All
students taking the questionnaire were told that they could stop any time they
wished and that it was not mandatory to complete the questionnaire once
started.

After obtaining the permission of the school principal, students in the
private school were contacted by a letter that explained the purpose of the
study. Letters were distributed to all the students in the junior and senior
classes. Copies of the letters to the juniors and seniors are found in Appendix
B. Students whose parents signed the letter and returned it to the school
participated in the study. The students who had parental permission were given
copies of the appropriate questionnaires by their homeroom teachers to be
completed during homeroom period. Completed questionnaires were taken to
the office by the teachers and were picked up by the author the next day.

The author introduced the study to the seniors in the public school during
a senior assembly. (For introduction to the study see Appendix C).
Participation was taken as student assent per agreement with the school
principal. Questionnaires were distributed to the seniors wishing to participate
and they completed the questionnaires during the period. Those who were not

finished returned their packets to the author at the end of the next period.

48

The study was explained to the juniors in their home rooms by the home
room teachers. Again, participation was taken as student assent. In addition to
the instructions on the questionnaire, the juniors were told who was doing the
study and the purpose. Due to a change in procedure for distributing the
questionnaires to the juniors, they were not instructed to write their sex on the
booklet. Teachers distributed the questionnaires and collected them as soon as
the respondents were finished. The author picked up the packets from the
school as arranged with the principal.

All questionnaires were distributed and completed within the last three
weeks of the school year. For the public school it was 2 1/2 weeks prior to
graduation; for the private school it was two weeks prior to graduation. The
end of the school year was chosen because the focus of all the seniors is on
graduation and bringing their high school days to an end at that time. Also, it
seemed to be a time when the other students in the school would be aware of it
being "the end" for the seniors. According to the high school teachers and
counselors with whom I talked, students realize that their high school days are
coming to an end at different times during the year (e.g., at the last football or
basketball game, the last play, the last band competition, or the last choir
concert), but graduation brings this back into focus for them as well as for

those students who did not participate in extra-curricular activities.

49

Less than one percent of the questionnaires given to the seniors were not
completed; none of the questionnaires given to juniors had to be discarded
because they were not finished. Questionnaires that were not used in the
analyses were those in which questions were not responded to item by item.
This involved 37% of the questionnaires that were given to the seniors and 13 %

of the questionnaires given to the juniors.

RESULTS

As is often the case with questionnaire research, some respondents chose
not to respond to all items. There were 281 seniors who responded to all items
and were included in the regression analysis, correlations and t-tests. There
were 291 juniors in the study; 55 responded to all items and provided necessary
information.

Xminfighml

Hypothesis 1 stated that high school seniors will have higher scores on a
depression index (CES-D) than will high school juniors. The data appropriate
for testing hypothesis 1 consist of the subject’s CBS-D scores. Each score is an
indication of the number and frequency of depressive symptoms experienced by
the individual. The mean CBS-D score for high school juniors was 21.46
(s.d. = 9.89). Seniors had a mean CES-D score of 18.96 (s.d. =10.83). The
difference between these means was significant [F(1,570)=4.82, o< .05].

Since the mean score of the juniors was significantly higher than that of the
seniors, the original hypothesis was not confirmed. In fact it is

counterindicated.

50

51

swarm

Hypotheses 2 and 3 focus on the impact of previous separations on
depression. Hypothesis 2 states that the more separations a high school senior
has had the higher will be his/her CES-D score. (Recall that the separation
history index is the sum of separations the respondent marked on a checklist.)
The number of separations checked ranges from 0 (one subject) to 9 (three
subjects). As seen in Table 1, there was a significant, negative correlation
between separation history and CES-D, or depression score, (r=-.12, p< .05).
Although the correlation was weak, it suggested that the more separations a
subject had experienced the lower the depression score was likely to be.

Hypothesis 3 states that seniors who had experienced the death of a
family member or close fiiend before the age of seven will have higher CES-D
scores. The prediction was not supported. The observed relationship between
death of a family member or close fiiend before the age of seven and
depression score was r= .06, p=.16. Because the sample provided too few
cases (just one senior reported the death of an immediate family member before
the age of seven and three more reported the death of a significant other during
their pre-school years) no conclusions can be drawn. Including the entire life
span (0-18 years) did not increase the number of instances significantly.

A step-wise multiple regression analysis (Table 2) indicated that total

number of separations contributed 1% to the variance (p< .05). When total

90:52.8 053v

 

_o.- *3 .- mo.- No. cor - - i -- -.. :Um:<
”Begum
9.58582
*0—.- NO. NO. ***MN.- *anNK 86 I - - - m0 ”tom
.8 8.58
NOV *m_ . we. cor 8. mo. unm— .- i - - mun—m
”acacia 3
326220“
nor 5. nor mo. cor ***0_. ***w_.- - - -- mm
”£26223—
baa...
fl

***m©. ***Nm. vo. *N—. we. 00f Nb.- MON - -- AMm ”roam:—
53383
No. scam— .- ho. mo. NOV 8. 8. **©_ .- “L _.- - xom
cor cor ***o_. —o.- 006 8r Sibm. scrim... “gm—.- xnb—N Q-mm—U
aofimoaoa
a @- E0 3 NJ .._-_.< a MW mud flaw 33.3”?»

= a t

_ 2an

 

:5. Vania." Ac. Vans." .mc. Va...
£53 8.. Bacon—cog Ho: macaw—obou ”mad

 

53

It It It II it II II it It It It 0
”83338
$50

***om.- I I I I I I I I I I n—
”58u 9
26 neonauaqom

no. No. - -- -- -- -- - - - .- E0
”82.5%

30> on 8

can.» in: 809nm

***NN. I I I I I I I I mn—
noEe: Eat

9»: 9 mam—q

80.33 .5 33.—

3.23% . *uamN. I I I I I I I >4
who» :8:

2.: 2 mum—n
.838 823

a @- .ED. 3 .3: .E.< Md Im- m-m. Iml glow Elana-N

a:
m
D.-
to

 

Goa—.950 2%:

54

34%

soda

coda
oNSN
ovdm
mhdv
36m

3.3

flag

**mm.m

$3.3

mod

shod

*wm.m
$3.8
$3.2

i262

mad-

3.0

No.0-
:6-
Id-
9 .o-
mad-

ow.

aluom

wfim-

ow;

S .o-
mNd-
owd-
Rd-
E .m-

34

8.

mm.

5.
5.
No.
8.

ma.

mm.

mm.

mm.

mm.

NM.

Tm.

wN.

mm.

xom

row we 8E8
a £222

Begum 2 H=QE=Q§<
acacia 9 383.20%
@055 “Bowman—om
amen—530m beam
xom

20m 3 .538
a £322

3

 

 

N 2.3

 

.—O. V9.3." .mo. VG...

.m©.— H = mmm%_d:< MOM .No.— I." — mmmhac< how cogs; Cmefln ”ouoz

 

55

v0.2
2.:
792
SQN
mm.w~
vaN
26m

mbdv

Hg

cod

and

cod

cod

*cw.m

*cm.w

*OhN

**mm.w

vod-

Ed

36

Ed

bmd-

2.0-

2.0

N_.o-

an

mic-
v0;
2 .N
no;
mod-
omd-
oo.m

Ed-

8.

5.

No.

No.

No.

hm.

hm.

om.

0m.

0m.

cm.

mm.

_m.

Begum 8 808582
550 2.8%me .0 53m
539 $.63me coma—«now
Bop—om

.0535 coup—30m

mucotm 2 32.33—3—
owsgu a: 383?...

338m 3 $05223—

Hana

 

56

number of separations was broken down into death of significant others and
separations excluding death, neither alone accounted for a meaningful portion of
the CES-D variance.

Table 3 presents each item from the separation history question and the
percent of senior subjects who checked that item as true. In addition to the
data presented, one subject had experienced the divorce of parents followed by

death of a stepparent.

 

 

Table 3
' r’ ' Hi It

item Rem
Lived in another town before living here 52%
Spent a week or longer away from family 90%
Attended nursery school or day care before

kindergarten 51 %
Living with one natural parent due to divorce,

separation, etc. (other than death) 26 %
Had experienced death of a parent 4%
Had traveled out of home state 98 %

Had experienced death of a close friend 4%

 

 

57

E . . 1 Cl

Hypothesis 4, the prediction of a significant positive relationship between
high school seniors’ indicating anticipation of more substantive changes next
year (Future Plans question 7) and the CES-D score, was supported (r=.19,
p< .001). Ninety-six percent said that next year their lives would be very
different from this (their senior) year. An expanded step-wise multiple
regression analysis indicates that the expectation that the year following
graduation would be very different from the senior year contributes 2% to the
CBS-D variance, p< .05, (Table 2).
Wm

It was predicted in hypothesis 5 that high school seniors with higher
Family Relatedness scores would have lower CES-D scores. Family
relatedness was negatively correlated with CES-D score (r=-.21, p< .001)
suggesting that a stronger relationship with family was associated with fewer
depressive symptoms. This supported hypothesis 5. The mean Family
Relatedness score was 25.56 (s.d. =6.01). As seen in Table 4, senior women
scored significantly higher than did senior men. A step-wise multiple
regression analysis (Table 2) indicated that family relatedness contributed

significantly (p< .01) to the CES-D variance.

Sex Differences

58

Hypothesis 6 states that females will have higher CES-D scores than

males. The data appropriate for testing hypothesis 6 consist of the subject’s

CBS—D scores. Each score is an indication of the number and frequency of

depressive symptoms experienced by the individual. Since there was no

 

 

Table 4
-an -anq. . Dvi 'n an: -v fr nior ’ R‘wlatne s V 'ole
Mable: 5% E M 141. um
Family Relatedness Male 112 24.41 5.98
2.63**
Female 169 26.32 5.93
Total 281 25.60 6.01
Relatedness to
Friends Male 112 32.02 5.74
213*
Female 169 33.34 3.97
Total 281 32.91 4.79

 

*p<.05. **p<.01.

59
significant difference in (CES-D scores) between schools [F(1,570)=0.07, NS],

further analyses were based on data from the two schools combined. (Due to
an oversight, gender information on the public school juniors was not obtained
and they were not included in the analysis.) The ANOVA revealed a
significant main effect for sex [F(l,330)= 19.58, p< .001] and a significant sex
by year interaction [F(l, 330)=5 .17, p< .05]. Post hoc t-tests indicate that
senior women scored significantly higher on the CBS-D than did senior men,
t(279) = 2.60, p< .01, and junior women scored significantly higher on the
CES-D than did junior men, t(53) = 2.18, p< .05. These findings support
much of the previous work showing more depressive symptoms among women
than men. (Table 5 presents the means.) The interaction suggests the
difference between CES-D scores of junior and senior women was greater than
the difference between the CES-D scores of the junior and senior men. This
sex by grade interaction was not predicted and may reflect the fact that there
were significantly more senior men and women in the sample than junior men
and women. Because of the large difference in sample size between years, the
results should be interpreted with caution.

Although school was not a significant variable it is interesting to note
that the female students in both the public and private schools scored higher on

the CBS-D than did the male students.

60

 

 

 

Table 5
r 1 n r E -D x n Y
lea; & Mean .N_
Junior Male 18.40 22
Female 24.91 33
Senior Male 16.88 1 13
Female 20.33 168
AttammenLtSLScthLl

Hypothesis 7 states that seniors’ CBS-D scores will be positively related
to their reporting that they will miss their high school (question 11 of the
Graduation and Future Plans questionnaire). The relationship between seniors’
tendency to agree with the statement "I think I’m going to miss this school very
much" and their CES-D scores was not significant (r=-.06). Hypothesis 7 was
not supported. Thirty-seven percent of the respondents said that they would
miss their high school very much. Table 6 presents this information. It is
interesting to note that 56% of the private school students agreed or strongly

agreed with question 11 but just 28% of the public school students agreed or

61

 

 

Table 6
r 0 mn- o l . r o'o.nndF ir-Pln O -os'ns
9925119.!) Agree Dis e
1. I am happy to be graduating. 98% 2%
2. I am sad about graduating. 40% 60%
4. I know what I am going to do next
year. 87 % l3 %
5. I am happy with the plans that I have
made for next year. 90% 10%
6. I know what career I am going to have. 96% 4%
7. I think next year my life will be very
different from this year. 96% 4%
9. I have made plans to see my fiiends
every other week next year. 58 % 42%
10. I have been active in school clubs
and organizations. 48% 52%
11. I think I am going to miss this
school very much. 37% 63%

 

Bette, Breakdown by public and private school for questions 10 and 11 is as
follows: 10. Public 45% Agree; Private 57% Agree. 11. Public 28% Agree;

Private 56% Agree.

62
strongly agreed. A post hoc Chi square analysis indicates the public/private

school difference is significant, X’z(1, N =281)=l9.76, p< .001. There were
no significant sex differences with respect to student’s anticipation of missing
their school.

Forty-eight percent of the seniors saw themselves as having been active
in school clubs or organizations (question 10), although 57% of the respondents
of the private school agreed as compared to 45 % of those from the large
consolidated public school.

AdditionaLAualyses

913M Because self-esteem has been a significant factor in
previous studies of depression, the Rosenberg Self-Esteem Scale was used as
the Opinion of Self index. It was given to all seniors. Low scores indicate a
better opinion and high scores indicate a lower opinion of self. The CBS-D
scores of the current subjects correlated significantly with Opinion of Self
(r=.50, p< .001). This is consistent with previous research. A step-wise
multiple regression analysis revealed that, of the six predictors, opinion of self
emerged as the variable making the most crucial contribution to the variance
(R2=.25, R2 change=.25) or 25% (Table 2). Inspection of the data indicates
that males in the private school had the most positive self opinion (M=1.73)

whereas males in the public school had the most negative opinion of themselves

63
(M=2.61); females from both schools fall in between. After controlling for

self-esteem, the remaining variables accounted for 8% of the CBS-D variance.

W The Relatedness to Friends index was given to
all subjects. The mean for seniors was 32.91, (s.d. =4.99). (The mean for
juniors was 31.12, s.d. =5 .20.) The senior women scored significantly higher
than did senior men, t(279)=2.29, p< .05 . (See Table 4.) There was a
significant negative correlation (r=-.l6, p< .01) between CES-D scores and
Relatedness to Friends.

W In this study 50.1% of the seniors scored 16 or more
on the CES-D, a score that indicates clinical depression. However, Roberts et
al. (1991) suggest a CBS-D score of 24 is optimal for determining clinical
depression in adolescent populations. Using this criterion at least 27.1% of the
seniors and 34% of all subjects in this study reported depressive symptoms at a
frequency or intensity to suggest clinical depression. Roberts et al. (1991) did
not report adjusted percentages using 24 as the criterion for clinical depression.
Therefore, a comparison of percentages between studies is not possible.

mm The percentages of the responses to the question about
where the students plan to live next year are presented in Table 7. Given the
large percentage of students planning to go to college it is not surprising that so
many will be in other towns. The nearby state university is a commuter

campus but some of those students get apartments near campus and some of the

64

 

 

Table 7
iv' n
m 2am
Plans to live at home 22.6%
Plans to have own apartment 14.2%
Plans to live in another town 61.9%
Unknown 1.1 %

 

seniors had indicated this option. The correlation between choice of residence
next year and family relatedness was not significant. Plans to live away from
home correlated significantly with perception of next year as very different
(r=.80, p< .001).

As indicated above, the subjects in the current study had a variety of
options. If the choice was to move away from home there was some ability to
control how far away from home one moved; thus a question was included to
find out how far away from home the seniors planned to go. Table 8
summarizes these results indicating the percent of the seniors choosing each
option. The correlation between number of miles away from home and family

relatedness was not significant (r= .05).

65

Table 9 shows seniors’ plans for the next year. Given the overwhelming
majority of middle class students in both schools the high number planning on
college is not surprising. Also, participation in the project was voluntary.
Therefore, caution is needed in generalizing the results to other socio-economic
groups.

Table 6 presents the percent of students who agreed and disagreed with
selected questions from the Graduation and Future Plans section of the
questionnaire. It is interesting that some subjects agreed that they were happy

about graduating and sad also.

 

Table 8

Distanee Seniers Plan 19 Live frem Heme

 

Item Emil
Live at home 23%
1-50 miles 12.6%
50-100 miles 19%

100 or more miles 45.4%

 

 

66

 

Table 9

Plnr NY

 

Item
None indicated
Military Service
Working (full or part time with no
other plans

Continuing Education

Percent

3%

5%

12%

79%

 

mete, Marriage was indicated by one individual who listed no other plans.

DISCUSSION

Several interesting findings emerge from the present study. It is
apparent that at the end of the school year many high school juniors and seniors
report a variety of depressive symptoms occurring more often than once or
twice a week. Several of the predictors suggested by the hypotheses of the
current study contribute to the variance in respondents’ reports of depression
related symptoms. The most significant contribution to depression related
symptoms is opinion of self. This finding supports the results of previous
research. Separation history contributed slightly to the variance of reported
symptomatology also. Although the data do not support all of the original
hypotheses the results are suggestive for further research. In this chapter
relationships among the CBS-D score, Opinion of Self, Family Relatedness,
Relatedness to Friends, Separation History, sex, year in school, feelings about
high school and plans for next year are explored. The nature of these findings
as well as some possible interpretations and implications are considered.
12mm

Contrary to what was predicted in hypothesis 1, high school juniors
report significantly more depressive symptoms on the CES—D than do the high

67

68

school seniors. Although most previous research (Allgood-Merten, Lewinsohn
and Hops, 1990; Roberts, Lewinsohn and Seeley, 1991; Roberts, Andrews,
Lewinsohn and Hops, 1990) had not found significant CES-D score differences
between high school age groups, Doerfler, Felner, Rowlison, Raley and Evans
(1988) reported CBS-D means from a sample of fourth through twelfth grade
students and found the mean CES-D increased through the eighth grade and
then began to decrease with a statistically insignificant increase noted for
eleventh grade students. R.D. Felner (Personal Communication July 31, 1991)
reported that the CES-D was given during the spring term of the school year.
The current study, also done during the spring, supports the finding of Doerfler
et al., 1988, i.e., the mean junior CES-D score was higher than the mean
senior CES-D score. Perhaps something happens during the junior year that
results in a relatively higher frequency of depressive symptoms.

The assumption in hypothesis 1 is that the period of time immediately
preceding graduation will be more traumatic for seniors than juniors because
seniors will experience multiple separation. They will be separating from their
younger friends, their senior friends, teachers, school and often from their
families. It is consistent with Bowlby’s, theory (1969) that the nature of the
attachment to family changes during adolescence and throughout life. He
suggests that the majority of adolescents develop attachments to friends,

teachers, groups and institutions (such as schools) as well as to their families.

69

The strength of non-family attachment is realized when relationships end or a
fiiend is not readily available. Weiss (1982) reports that adolescents often seem
surprised by the strength of loneliness and distress they feel when experiencing
separation. He suggests that adolescents are surprised by their negative feelings
because there is an implicit understanding that the relationships of adolescence
are transitory. The implication is that the relationships are not as serious as
their feelings ultimately indicate they are. According to Bowlby (1969), if an
attachment has developed separation from individuals, groups or institutions
will activate the self-in—separation model and trigger the feelings therein.
Although activating the model would not lead to depression in all seniors, the
multiple separations make it seem likely that whatever negative feelings are
associated with separation would be triggered.

One possible explanation for the higher CES-D scores of juniors may be
that the juniors will be losing their senior friends and, although this kind of
separation is not new (it occurred in elementary school and middle school), this
time it is a permanent separation for most of them. Some support for this
position may come from an unusual study by Field (1984). Field studied
children who were to be ”left behind” as some of their friends and classmates
were being transferred to another nursery school. The pre-school subjects in
her study who were going to remain in their current school showed symptoms

of depression prior to the separation whereas those who were moving showed

70

few symptoms before and more symptoms after the move. Granted there is a
large age difference between Field’s pre-school subjects and the juniors in the
current study, yet their responses could be seen as similar. In fact, it would
seem that the juniors in the current study are more capable of anticipating their
loss than the pre-schoolers. Cognitions such as those involved in anticipation
can trigger emotional reactions (Ellis, 1968; Maultsby, 1980).

A second possible explanation of the juniors higher CES-D scores is that
the juniors begin to realize that they will have to assume leadership roles.
Concern about competence may be activated by those confronted with the
responsibility of leadership. They may be concerned, too, that they will not be
as competent as the seniors are. On the other hand, the juniors may look
forward to the prestige associated with leadership and the explanation for their
higher CES-D scores may lie elsewhere.

Another explanation is that the juniors are experiencing stress that
exacerbates a depressed mood resulting in CES-D scores that are higher than
the seniors’ CES-D scores. It is generally accepted that taking tests and being
interviewed are stressful situations. If we assume that most of the juniors in the
study, like most of the seniors, are college bound they may be experiencing
anxiety over SAT and ACT test performance and about being accepted by the
college of their choice. Allgood-Merten et al. (1990) found that stressful recent

events were associated with higher CES-D scores. There is no independent

71

measure of anxiety, stress, or of juniors’ feelings about their senior year. This
suggests an area for further research.

Several factors may moderate the reaction to graduation for seniors.
Seniors’ plans for the coming year are complete (97%) and 90% of the seniors
in this study report "I am happy with the plans I have made for next year."
Focusing on the future and the advantages of graduation may moderate the
negative aspects of loss associated with graduation and account for the grade
difference in depressive symptomatology. The fact that the seniors are happy
with the plans they’ve made for the coming year may be related to the fact that
98% of them said they were happy to be graduating. However, happiness is
not the only emotion the seniors are feeling because 40% of the seniors said
they were sad (about graduating) also.

Seamfieaflistm

Rather than the positive relationship between the number of previous
separations and CES-D score that was predicted in hypothesis 2, there is a
- negative relationship. The results indicate that students who have experienced
more separations report fewer depressive symptoms.

Bowlby (1969) suggested that current separation experiences might re-
activate the intemal self-in-separation model. The content of the model can be
unconscious. When the model is activated, individuals who have experienced a

number of stressful separations in childhood (e. g., parental death, parental

72

divorce, loss of friends and/or loss of home due to a move) experience the
negative emotions again. Studies by Bloom (1977/1978), Podell (1989) and
Stein, Levy and Glasberg (1974) seem to support Bowlby. Some of the seniors
in this study reported feeling both happy and sad. The sadness they feel may
be a result of the activation of the negative feelings in the separation model.
Another possibility is that the separation model is modified by
subsequent experiences in which separations are dealt with successfully. The
result would be an internal working model of self—as-competent—in-separation
situations. Those seniors who have coped successfully with previous
separations may have integrated these experiences into their self-in—separation
model. The revised model includes a competent, coping self, i.e., one that
finds opportunities or expects positive outcomes from separations. Some of the
experiences marked most frequently by the subjects, e. g., vacationing away
from home, spending a week or more away from family, moving and nursery
school attendance may modify the original separation model positively.
Divorce or separation of the parents could work either way. In a study by
Proulx and Koulack (1987) adolescents reported feeling more control in
situations where the divorce related conflict was in the open than in those
situations in which conflict was not open. Moore and Hotch (1982) found a
significant personal control variable in their studies of the effect of divorce on

older adolescents’ perception of having left home. Feeling in control can

73

support feelings of competence and expectations of positive outcomes. The
more separations one has mastered the more confidence there is to handle new
separations. This is a positive, growth-oriented perspective and not contra-
indicated by Bowlby’s work. A practical application that is supported by the
negative relationship between number of separations and depressive symptoms
is that children and adolescents do not need to be shielded from separation
experiences. What is needed is research of the self-as-competent-in-separations
model.

A third explanation is that more recent events have a greater impact on
depression than past events such as separations. That is, the correlation may be
due to graduation or to a factor not evaluated in the current study. Allgood-
Merten et al., (1990) found that ”more distant, low base-rate major events"
correlate less strongly with depression than do recent life events. It is possible
that the negative relationship found here is due to the immediacy of graduation
and that concerns and issues relating to graduation and plans for the future
override past events.

Hypothesis 3 stated that seniors who had experienced the death of a
family member or close fiiend before the age of 7 would have higher CES—D
scores. Just one student had experienced such a loss. This student had lost a
parent. There were ten seniors who had experienced the death of a parent and

two who had lost a sibling. To determine the long-term impact of stressful or

74

traumatic losses such as death within the family, a larger sample or a study
with matched controls is needed.
An ' i . h n

Hypothesis 4 stated that scores on the CES—D would be positively related
to the anticipation that the coming year will be substantially different from the
senior year. The hypothesis was supported. Ninety-six percent of the seniors
expect that the coming year will be very different from their senior year. Plans
to live away from home were significantly correlated with seniors’ perception
that the coming year would be different from their senior year in high school.
The anticipated difference between years must be based on factors other than
the role of student since many of them will continue (to be students. The
anticipation that their lives would be very different suggests that graduation may
be of psychological importance as a marker of separation from the past. It
appears that moving away from the family is a major factor. The relationship
between adolescent individuation and physical distance from parents is
suggested by the work of Sullivan and Sullivan (1980).
W

The results of this study supported the negative relationship between
family relatedness and CES-D score predicted in hypothesis 5. These data tend
to support the work of Kandel and Davies (1982) who found that parent

closeness was associated with lower depression score among older adolescents.

75

(Kandel and Davies measured depression with a six item depressive-mood
scale.)

Feeling close to one’s family may be experienced as positive or negative.
The family can be experienced as supportive and as a "secure base." On the
other hand, the family may be experienced as negative if the adolescent feels
threatened with being smothered or enmeshed. It is possible that the depressive
symptoms reported by adolescents reflect an ongoing struggle between the need
for a secure base and a concern about being enmeshed and not being able to
become independent.

Adolescents are seen as working toward separation from their families.
Separation can be physical, psychological or both. Psychological separation is
referred to in the literature as individuation (Blos, 1961; Josselson, 1980;
Sabatelli and Mazor, 1985; Allison and Sabatelli, 1988 among others). It is
linked to a sense of autonomy and personal identity. Individuation does not
mean that there is lack of connectedness to the family. Rather, it involves an
acceptance of the ways in which one is similar to and different from ones’
family. It is a sense of belonging to but not feeling enmeshed in the family.
Individuated adolescents feel close to and consult with their parents without
feeling psychologically threatened. Greenberg et al. (1983), reported that the

majority of adolescents in their study felt this way.

76

Seniors may feel individuated, having made some important decisions
about their future. Montemayor (1982), Youniss and Smoller (1985), and
Newman (1989) found that parents of 17 and 18-year-olds expect more
independent decision making. Allison and Sabatelli (1988) reported that
adolescents living in healthy, well differentiated families were likely to have
freedom and opportunities to make their own decisions. Independent decision
making gives seniors control over their own lives. Those who did not have as
much say in the decisions made about the next year may see graduation as the
beginning of an opportunity to make their own choices.

High school juniors, as a group, may be affected by graduation because
they are very involved in the struggle toward individuation. Many juniors will
have not made final decisions about their future. They may be in conflict with
their families over options and feel the conflict more acutely at the end of the
junior year. There was no measure of individuation in this study. Thus this
explanation is speculative and suggests an area for further research.

Bowlby (1973) says that it is important for an individual of any age to
feel "confident that an attachment figure will be available to him whenever he
desires it" (p. 202). The attachment figures and family serve as a "secure base"
from which the individual gradually leaves to explore the world (Bowlby, 1988,
p.11). When attachment has been positive the individual ventures away,

periodically checking back and returning if frightened or if something goes

77

wrong. In these situations the distance between the individual and the
attachment person is controlled by the individual. Physical contact is necessary
to relieve separation anxiety for most young children. With increased maturity
visual and verbal contact can substitute for physical contact. Telephones enable
individuals to maintain verbal contact with attachment figures. In addition to
control over access to others by telephone, the high school seniors in this
midwestern town have a variety of options for controlling the distance they
wish to venture from home base. There are two universities (one state and one
private) and several technical schools within 30 minutes of their high schools.
There are numerous colleges, universities and technical schools within 50 to
180 miles. Having such choices may attenuate the perception of graduation as
separation for those needing their secure base a little longer or needing to
return to it more often. Adolescents who feel secure in their family
relationships may venture further from home base after graduation whereas
those with more conflicted relationships may stay closer to home. Of the 281
seniors, 76.5% planned to live some place other than with their parents and
61.9% planned to live in another town. Information relevant to the nature of
relationships, secure or conflicted, is not available for the subjects in this study.
This could be the subject of another study.

Being able to access the secure base is another way of perceiving the

issue of control. It has surfaced in a variety of studies of separation.

78

According to Irion et al. (1988), children from intact families had greater
feelings of control and viewed stressful situations as opportunities more often
than did those from disrupted families. Proulx and Koulack (1987) also
reported that feelings of control in stressful situations were more common
among college students from intact families than from disrupted families. The
observed negative relationship between CBS-D scores and family relatedness
may be a reflection of the fact that nearly three—fourths of the seniors in this
study are from intact families.

Relatednesflejdends, A positive correlation between family relatedness
and relatedness to friends lends support to Bowlby’s position that positive
family relationships make possible the development of friendships and
attachments to others as the child grows.

In the current study lower CES-D scores were associated with higher
relatedness to fiiends scores. A possible explanation for students having higher
relatedness to friends scores and fewer depressive symptoms may be that
friends have become available attachment figures and provide a secure base
similar to that provided by the family (Bowlby, 1973). The importance of
maintaining contact with fiiends may be seen in their response to the statement
"I have made plans to see my friends every other week next year"; fifty-eight

percent of the seniors agreed or strongly agreed with the statement.

79

The statement "I have made plans. . . ." implies some control and some
independence. There was a significant relationship between a more positive
opinion of self and higher relatedness to friends score. The seniors may feel
independent and confident (positive about themselves). They may believe that
they can maintain current friendships and develop friendships in new settings.
Sex Differences

Hypothesis 6 states that females will have higher CES-D scores than
males. This sex difference was supported for both juniors and seniors. In this
respect the results of the current study are consistent with the findings of Rutter
(1986b), Allgood-Merten et al. (1990), Ehrenberg, Cox and Koopman (1990),
Roberts et al. (1990) and Roberts et al. (1991). The sexual stereotype in this
society is that women usually acknowledge their emotional reactions while men
tend to suppress their emotional reactions to separation. It is more likely that
men and women express their feelings differently. Bowlby (1980) discusses a
possible basis of the inability to express feelings; he makes no distinction based
on gender. He says that individuals who grow up in families that do not value
attachment bonds and that consider the expression of feelings and attachment
behaviors a weakness " grow up tough and hard" (p. 225). It is possible that
the relatively lower male CES-D scores are a result of a cultural prohibition

regarding the freedom of men to express their feelings related to attachment.

80

The difference between the sexes was more striking for the juniors than
for the seniors and accounts for the significant sex by year interaction in the
ANOVA of the depression scores. Few studies of depression in adolescence
report sex by age interactions. Rutter (1986b) reported a sex by age interaction
related to puberty. Allgood-Merten et al. (1990) and Ehrenberg, Cox and
Koopman (1990) reported trends that suggested possible sex by year
interactions.

The basis for the significant sex by year interaction is not entirely clear
nor was it predicted. However, some of the research on the relative
importance of relationships for men and women may shed some light on the
issue of a sex difference that interacts with year in school. Douvan and
Adelson (1966), Miller (1976), Fischer (1981), Gilligan (1982), and Frank,
Avery and Laman (1988) report that women are more interested in and
concerned about relationships than men. Rutter (1986a) indicated that the stress
of disrupted relationships can lead to feelings of depression, or depressed
mood. A possible explanation for females’ higher CES-D scores especially in
the senior year, is that females perceive graduation as disrupting relationships.
There would be more relationships disrupted for senior women than for junior
women.

Another possible explanation for the sex by year interaction might be that

the senior males are more career oriented and confident about such decisions

81

than are the senior females. The data from this study, however are insufficient
to draw any conclusions.

Roles may be another factor in the sex by year interaction. Being a
senior usually implies leadership. Junior males may be delighted that they will
finally have their chance to be leaders, a traditional male role. Females, on the
other hand may experience more anxiety in assuming leadership roles. The
females may perceive being leaders as a relationship issue. That is, leadership
means being in competition with the males who will be available for dates.

Further research on sex differences in the nature and meaning of
leadership among high school students may help clarify the role of the
relationship variable in the sex by year interaction in depression scores.

hm n h 1

Feeling that one is a part of school, whether through taking part in clubs
and activities or because of relationships with teachers or peers, could make
graduation seem more traumatic. Hypothesis 7 predicted a positive correlation
between attachment to school and CBS-D. This was not supported. Fewer
than half of the seniors reported feeling "attached" to their school in the sense
that they think they will miss it once they graduate. Fewer than half reported
having been very active in clubs or organizations. There is an interesting
difference between public school and private school students. Over half of the

seniors from the private school reported that they thought they would miss their

82
school and/or that they had been active in clubs. Slightly more than a quarter

of the public school seniors thought that they would miss their high school and
just under 50% of them had participated in clubs. Barker and Gump (1964)
have reported that students from smaller schools are more actively involved in
their schools. The current study lends support to their work.
Additionalhsues

W The CES-D means reported for
the normal adolescent subjects in the studies of Allgood-Merten et al. (1990),
Roberts et al. (1990), Doerfler et al. (1988) and in the current study (see Table
5) are higher than the reported mean CBS-D scores in studies of normal adult
samples. Using the adult caseness criterion, 12% to 50% of the subjects in
these studies would be labeled clinically depressed. Of the high school seniors
participating in this study nearly 50% would be labeled clinically depressed. It
is not clear what the higher high school means indicate. One explanation is that
adolescents are more depressed than adults. On the other hand, the fact that
adolescents report more symptoms of depression on the CES-D than do adults
does not necessarily mean they are clinically depressed. Breslau (1985), for
example, found that mothers of children with chronic disabilities had CES-D
scores indicative of depression but by other measures (of depression) no higher
level of clinical depression than the normal population. A third possibility is

that the higher prevalence of symptoms reflects normal adolescent turmoil.

83

Adolescents may exaggerate disappointments and react with more extreme
affect and drastic cognitions than do adults. Perhaps the more extreme
reactions to events makes the timing of assessment of depression or depressive
symptoms critical in studies of adolescents.

Roberts et al. (1991) suggest that a CES-D score of 24 is optimal for
determining "caseness" in adolescent populations. Using this criterion, fewer
than 27% of the seniors and 34% of all the subjects in this study are reporting
depressive symptoms at a frequency or intensity to suggest clinical depression.
These percentages seem to be large enough to suggest a need to continue to
investigate adolescent depression.

Since the CBS-D is designed to reflect current feelings rather than an
ongoing state, it is possible that different results would be obtained were it
possible to give the questionnaire to both juniors and seniors the day of
graduation. It would be interesting, too, to compare responses from the
beginning of the senior year with responses at the end of the year. The
difference score would be a more sensitive reflection of individual responses to
graduation.

ML The simple correlation between opinion of self and
CES-D score indicates that fewer depressive symptoms are associated with a
positive view of self. (The Opinion of Self scale was scored such that a low

score reflected a more positive view of self.) These results were expected and

84
parallel those of Doerfler et al. (1988) and Allgood-Merten et al. (1990) among

others. The opinion of self/self-esteem variable was included for that reason.

Conclusions

Although several of the hypotheses of the current study were not
supported, some interesting findings did emerge. Perhaps the expected positive
relationship between previous separation experiences and responses to current
separation experiences would be found had the research been carried out
differently. A number of limitations and implications for further research exist.
Extent ef Implieations ef the Qgrrent Study

There are several cautions regarding the generalization of the results of
the current study. As described in the methods section the subjects in this
study were primarily middle class, white, midwestern high school juniors and
seniors. Further research with subjects of other socio-economic and racial
backgrounds and with plans other than to attend college is needed.

Likert type scales, such as those used throughout the questionnaire, are
subject to response bias. However, review of the individual questionnaires
suggests response bias was not a factor in this research. Only on the

Relatedness to Friends section of the questionnaire did there seem to be a

tendency to respond W (or W to 70% or more of

86

some children parental divorce may be a relief and seen as positive. For some
children of divorce arrangements may have been made so that separation of
parent and child was minimal, whereas for others parental divorce or separation
may have resulted in prolonged separation from an attachment figure.

Although this study is based on the assumption that graduation is a
separation experience, an independent measure of the student’s interpretation of
graduation was not obtained. This, too, could be seen as a limitation.
W

A variety of further studies are suggested by the current research. One
such study might involve assessing the presence and content of the internal
models proposed by Bowlby to explain responses to separation.

Second, the higher CES-D scores for juniors was not predicted. It
would be interesting to investigate what is happening in the lives of juniors at
the end of the junior year that is related in depression. One version of this
might compare CES—D scores of juniors applying to colleges and taking the
SAT and ACT tests with those who are not. Another version might compare
student’s CES-D scores during the time they are applying to colleges and taking
tests with the CES-D scores of the same students after college decisions are
made.

Third, an alternative design for assessing senior’s response to graduation

might involve administering the CES-D or similar instrument in the fall of the

87

senior year and comparing those responses with answers to the same measure
obtained in the late spring when graduation is imminent.

Another study that is suggested involves adolescents’, both juniors’ and
seniors’, perception of graduation. An open-ended questionnaire or interview
in which they respond to questions such as: What does graduation mean to
you? and How do you feel about graduating (graduation)? could be used.

The nature of the adolescent-parent relationship and the adolescent’s
perception of graduation as well as future plans could be investigated in more
detail. Groups of adolescents with secure vs. conflicted relationships (or
relationships in which the adolescent has achieved individuation vs.
relationships in which the adolescent has not) could be compared regarding
responses to current separations such as graduation.

A study could compare the relative impact of various separations (recent
and remote, mild and severe - as perceived by the individual) on adolescents’
reactions to a current separation experience.

Early parental divorce or death may have a long-term impact on
adolescents’ reactions to separations. The sample of adolescents experiencing
death of a significant other before the age of seven was too small, in this study,
to do an analysis. Therefore, a study in which subjects are selected and
matched on the basis of having or not having experienced a death before the

age of seven is suggested. A similar study investigating variables relevant to

88

children who have experienced parental divorce is suggested as well. Variables
of interest might include whether the child ended up living with the primary
attachment figure or not, the age of the child at the time of the initial
separation/divorce, whether or not the child was given understanding and
assistance in learning to cope with the separation, and the accessibility of the
non-custodial parent.

In summary, high school seniors report depressive symptoms within the
last 2-3 weeks of the school year. High school junior females report even
stronger feelings and behavioral symptoms of depression than do seniors
(female or male). As has been true in other studies, self-esteem or a positive
opinion of self is significantly related to fewer feelings associated with
depression. Opinion of Self score accounted for 25% of the variance in the
depression score. Other significant factors were sex, family relatedness and
number of separations experienced from early childhood to the present. The
data tend to support a higher rate of feelings of depression among females.
Fewer symptoms of depression are reported when there is greater family and
friend relatedness. Some seniors report feeling both happy and sad about

graduating.

 

APPENDICES

APPENDIX A

JUNIORS’ QUESTIONNAIRE INSTRUCTIONS

STUDENT QUESTIONNAIRE
(Juniors)

This is part of a study in which we are interested in what high school
students think about themselves, their friends and family and the future. We
are hoping to be able to better understand how they come to make their
decisions and form their ideas. These questions will ask for your opinions about
your friends and yourself.

There are no right or wrong answers to any questions. This is not a test
at all. We will not know your name. Anything you write is completely
private, so you should feel free to answer the questions as honestly as you can.

Please answer the questions in order. Try notto spend too long on any
one question.

Thank you very much for helping in this research.

DIRECTIONS

For each question or part of a question mark one answer. Do not write
your name on any of the materials. You may start immediately and continue
through the parts of the questionnaire until you have finished. If at some point
you decide that you do not wish to answer a question you may go on to the
next one. If you do not wish to complete the questionnaire, stop and turn in
the questionnaire to your teacher as it is.

89

 

SENIORS’ QUESTIONNAIRE INSTRUCTIONS

STUDENT QUESTIONNAIRE
(Seniors)

In this study we are interested in finding out about high school seniors
and about their plans and thoughts about graduation and the future. We are
hoping to be able to better understand how people come to make their decisions
and form their ideas. Some of the questions will ask for your opinions, while
others will ask for general information about yourself.

There are no right or wrong answers to any questions. This is not a test
at all. We will not know your name. Anything you write is completely
private, so you should feel free to answer the questions as honestly as you can.
Participation is completely voluntary. If at some point you do not wish to
answer any more questions, you may stop and return the questionnaire to your
teacher.

Please answer the questions in order. Try not to spend too long on any
one question.

Thank you very much for helping in this research

DIRECTIONS

For each question or part of a question, mark one answer, or write in the
information that is asked for. Do not write your name on any of the materials.
You may start immediately and continue through the parts of the questionnaire
until you have finished. If at some point you decide that you do not wish to
answer any more questions, you may stop and turn in the questionnaire to your
teacher.

90

91
GENERAL INFORMATION QUESTIONNAIRE

Please mark an X next to the correct answer
1. What is your sex? Male Female

2. What year are you in school Junior Senior

3. Who lives at home with you now? Put an X next to all who live with you

 

_Mother _Stepmother _Grandmother
__Father _Stepfather _Grandfather
__Sister(s) ___Stepsister(s) or halfsister(s)
_Brothers _Stepbrother(s) or half brother(s)

4. Next year I plan to _live with my family, __have
my own place _____live in another town

5. If you will be away from home, how far away will you be?

 

up to 50 miles 50-100 miles 100-200 miles
over 200 miles.
6. Next year I plan to be working full time and support-
ing my self in the military service going to

college or a technical school Other
(Please indicate)

7. Please check those of the following that apply to you.

I have lived in another town before living here.

I have spent a week or more away from my family at camp,
on vacation, etc.

I have an older brother or sister who lives away from home.

 

92

My parents are divorced or separated.

I went to day care or nursery school before I entered
school.

A member of my immediate family has died.

(Relationship: )
mother, father, sister, brother, stepparent

 

(Number of years ago: )

 

Someone else to whom I felt close has died.

(Relationship:

 

friend, grandparent, neighbor, cousin, etc

(Number of years ago:

 

I have traveled out of my home state.

I have traveled outside of the US.

 

93
GRADUATION AND FUTURE PLANS QUESTIONNAIRE

Circle one point on the scale that describes what your opinion.

SA =STRONGLY A =AGREE D =DISAGREE SD =STRONGLY

AGREE DISAGREE

1. I am happy that I will be graduating soon SA D
2. The thought of graduation makes me sad SA D
3. The thought of being on my own after

graduation makes me nervous SA D
4. I know what I will be doing next year SA D
5. I am happy with the plans I have for

next year SA D
6. I know what career I am going to have SA D
7. I think next year my life will be very

different from this year SA D
8. I worry that after graduation I will

probably never see most of my classmates

again SA D
9. I have one or two good friends that I

expect to see at least every week next year SA D
10. I have been very active in school clubs

and organizations. SA D
11. I think I’m going to miss this school

very much. SA D

SD

SD

SD

SD

SD

SD

SD

SD

SD

SD

SD

 

94
FRIENDS QUESTIONNAIRE

Please circle one point on the scale that follows each statement.

SA = STRONGLY A = AGREE D = DISAGREE SD = STRONGLY

AGREE DISAGREE

1. My friends are interested in what I do. SA D
2. I feel sure my fiiends like me. SA D
3. I talk with my fiiends about things

that are important to me. SA D
4. I think my relationship with my fiiends

is very close. SA D
5. I get along well with my friends. SA D
6. I would go to my friends if I were

feeling lonesome. SA D
7. I would go to my friend if I needed to

find someone who understands how I feel. SA D
8. When I am away from my friends I miss

their company very much. SA D
9. My friends are very important to me. SA D
10. Five years from now I expect most of my

friends will be people I don’t know now. SA D

SD

SD

SD

SD

SD

SD

SD

SD

SD

SD

APPENDIX B
LETTER TO JUNIORS

Michigan State University
E. Lansing, MI. 48824

Dear Parents and Students:

My name is Margaret Connolly. I am a psychology graduate student at
Michigan State University and am now living in Evansville. I am studying high
school students responses to graduation. As part of this study I need to find out
high school juniors ideas, opinions, and feelings at the end of their junior year.
There is a short questionnaire about friends also. I would like to know if you
would be interested in participating in the study during the last couple weeks of
the school year. The study is conducted under the guidance of Dr. John
McKinney and has been approved by Mr. Adams. It will require only 5 to 10
minutes of your time.

All that is asked is that you fill out a questionnaire. The questionnaire
will be given to you during class with the instructions not to put your name or
any identifying information on it. Participation is voluntary and if at some time
while you are working on the questionnaire you wish to stop, you may. Just
return the questionnaire packet to your teacher. I will pick up all
questionnaires after school is out the day you do it.

Please consider whether or not you would be interested in participating.
When you have decided to participate, please sign the attached form and return
it to the teacher from whom you received this letter.

Thank you.

95

96

I have voluntarily
decided to participate in the study that Margaret Connolly is conducting on high
school students ideas about graduation. I understand what is expected of me. I
understand that the questionnaire is anonymous and that I may withdraw from
the study at any time. I understand that I have the right to the general results
of the study which will be available to me by request. Participation in the
study does not guarantee any beneficial results.

 

 

 

Parent Date

 

Student

97
LETTER TO SENIORS

Michigan State University

Dear Parents and Students:

My name is Margaret Connolly. I am a psychology graduate student at
Michigan State University studying adolescents’ responses to graduation. I am
now living in Evansville. I am interested in what adolescents think about
graduating, how they feel with graduation close at hand and what effect
previous experiences of being away from or separated from their families has
on how they respond to graduation. I would like to know if you would be
interested in participating in the study during the last couple weeks of the
school year. The study is conducted under the guidance of Dr. John McKinney
and has been approved by Mr. Adams. It will require only 10 to 20 minutes of
your time.

All that is asked is that you fill out a questionnaire. There are
questions about your ideas and attitudes and feelings about your family, about
friends, about you, and about graduation. The questionnaire will be given to
you during class with the instructions not to put your name or any identifying
information on it. Participation is voluntary and if at some time while you are
working on the questionnaire you wish to stop, you may. Just return the
questionnaire packet to your teacher. I will pick up all questionnaires after
school is out the day you do it.

Please consider whether or not you would be interested in participating.
When you have decided to participate, please sign the attached form and return
it to the teacher from whom you received this letter.

Thank you.

98

I have voluntarily decided to
participate in the study that Margaret Connolly is conducting on high school
students ideas about graduation. I understand that everything in the study has
been explained to me. I understand that the questionnaires will be anonymous
and that I may withdraw from the study at any time. I understand that I have
the right to the general results of the study which will be available to me by
request. Participation in the study does not guarantee any beneficial results.

 

 

 

Parent Date

 

Student

99

APPENDIX C

INSTRUCTIONS TO PUBLIC SCHOOL SENIORS

(Introduction provided public school seniors at their assembly.)

Good morning.

My name is Margaret Connolly. I am a psychology graduate student at
Michigan State University studying adolescents’ responses to graduation. I am
now living in Evansville. I am interested in what adolescents think about
graduating, how they feel with graduation close at hand and what effect
previous experiences of being away from or separated from their families has
on how they respond to graduation. The study is conducted under the guidance
of Dr. John McKinney and has been approved by your school superintendent
and principal. It will require about 10 to 20 minutes of your time.

All that is asked is that you fill out a questionnaire. There are
questions about your ideas and attitudes and feelings about your family, about
friends, about yourself, and about graduation. The questionnaire will be given
to you in just a few minutes. Do not to put your name or any identifying
information on it. Participation is voluntary and if at some time while you are
working on the questionnaire you wish to stop, you may. Just return the
questionnaire packet to me by noon today. I will be sitting just outside the
cafeteria.

I hope you will be interested in participating.

Thank you.

 

 

 

LIST OF REFERENCES

LIST OF REFERENCES

Ainsworth, M.D.S. (1989). Attachments beyond infancy. American
PM. fl. 709-716.

Ainsworth, M.D.S., Blehar, M.,& Waters, E. (1978). Battemefattaehmem.
Hillsdale, N .J .: Wiley and Sons.

Allgood-Merten, B., Lewinsohn, P., & Hops, H. (1990). Sex differemces and
adolescent depression. WM 22, 55-63.

Allison, M.D. & Sabatelli, RM. (1988). Differentiation and individuation as
mediators of identity and intimacy in adolescence. LqemaLofAdelesgnt

Researeh. 3. 1-16.
Angold, M., Weissman, K., John, K., Merikangas, K., Prusoff, B.,

Wickromaratne, P., Gammon, G. & Warner, V. (1987). Parent and
child reports of depressive symptoms in children at low and high risk of

depression W. 28 901-915

Barker RUG &Gump. PV (1964) ELEMLMMMMDQ
WM. Stanford. Stanford Press.

Barnes, G.E. & Prosen, H. (1985). Parental death and depression. W
Aenetmallsxchiatrx. %. 64-69.

Beck, AT. (1970). W. Philadelphia: University

of Pennsylvania Press.

Bloom, M. V. (1977/1978). The death of childhood. The process of adolescent-
Parent separation W. 1978.32. 369-3

Blos, P. (1961). W. New York: Free Press.

100

101

Blos, P. (1967). The second individuation process of adolescence. In R.S.

Eissler, et al. (Eds), The psyeheanalyfle study ef the ehild Vol. 22 (pp.

162-186). New York: International Universities Press.

Blyth, D.A., Simmons, R.G., & Carlton-Ford, S. (1983). The adjustment of

early adolescents to school transitions. urn f E l A ole cence, 3,
105-120.

Bowlby, J. (1969). AW. New York: Basic
Books.

Bowlby, J. (1973). W. New York: Basic
Books.

Bowlby, J. (1980). Attgehment and loss; st. sadness and depressien. New
York: Basic Books.

Bowlby, J. (1982a). WWW (2nd ed). New York:
Basic Books.

Bowlby, J. ( 1982b). Attachment and loss: Retrospect and prospect. Amerieae
lorLrnaLoLthonsrLehiaer. 52. 664-678.

Bowlby,J. (1988). a ‘ r 0 ' ' r- s int n 2 0h... h
dexelegmfl New York: Basic Books.

Breslau, N. (1985). Depressive symptoms, major depression, and generalized
anxiety: A comparison of self reports on CES-D and results from

diagnostic interviews. W, 15, 219- 229.

Bretherton, I. (1985). Attachment theory. Retrospect and prospect. In I.
Bretherton & E. Waters. (Eds. ), Growing points of attachment theory and
research. n r h 1

W. 5.0. (1. 2. Serial No. 209).

Brown, G. W, Harris, T. O., & Bifulco, A. (1986). Long-term effects of early
loss of parent. In M. Rutter, C. Izard & P. Read. (Eds.,) Deming

W (pp. 251-296). New York: Guilford.

Carlson, C. & Cantwell, D. (1980). Unmasking masked depression in children
and adolescents. W. 1.31. 445-449.

102

Csikszentmihalyi, M. & Larson, R. (1984). Being adolescent; Confliet and
W New York: Basic Books.

Demo, D. (1985). The measurement of self—esteem: Refining our methods.

Jettrnal ef Persenalim and Seeial Esyehelegy, 48, 1490-1502.

Doerfler, L.A., Felner, R.D., Rowlison, R.T., Evans, E. & Raley, RA.
(1988). Depression in children and adolescents: A comparative analysis
of the utility and construct validity of two assessment measures. Jogrnal

ef Censttltjng and Clinieal Esyehelegy, 56, 769-772.

Douvan, E. & Adelson, J. (1966). W. New York:
Wiley and Sons.

Ehrenberg, M.F., Cox, D.N., & Koopman, RP. (1990). The prevalence of
depression in high school students. Wm, 25, 905-910.

Ellis, A. (1968). What really causes psychotherapeutic change? loiees, 4,
90-97.

Emery, G., Bedrosian, R. & Garber, J. (1983). Cognitive therapy with
depressed children and adolescents. In D.P. Cantwell & G.A.Carlson

(Eds.),ff'v 'rr'nhilh l n°An t
(pp. 445-471). New York: Spectrum.

Erikson, E. (1968). Identig; Yeeth and erisis. New York: W.W. Norton.

Field, T. (1984). Separation stress in young children transferring to new schools.
Whom. 29. 786-792.

Fischer, J .L. (1981). Transitions in relationship style from adolescence to young

adulthood. MW Ill. ”-23.

Fleming, S.J. & Adolph, R. (1986). Helping bereaved adolescents: Needs and

responses. In C.A. Corr & J.N. McNeil (Eds.), Adelesg‘enegandjeath
(pp. 97-118). New York: Springer.

Frank, S.J., Avery, C.B., & Laman, MS. (1988). Young adults’ perceptions
of their relationships with their parents: Individual differences in
connectedness, competence and emotional autonomy. Developmental

Esyebolozx. 24. 729-737.

103

Fried, M. (1963). Grieving for a lost home. In L]. Duehl and J. Powell

(Eds.), i'n°P 1 1i ' h m r li (pp.
151-171). New York: Clarion.

Frerichs, R.R., Aneshensel, C.S. & Clark, V.A. (1981). Prevalence of

depression in Los Angeles County. Ameriean Jeurnal of Epidemielegy,
113, 691-699.

Friedrich, W. Reams, R. & Jacobs, J. (1982). Depression and suicidal ideation
in early adolescents. f Y A le c n _1_1, 403-407.

Garmezy, N. (1986). Developmental aspects of children’s responses to the
stress of separation and loss. In M. Rutter, C. Izard & P. Reed (Eds.),

W (pp. 297-323). New York: Guilford.

Gibbs, J .T (1985). Psychosocial factors associated with depression in urban
adolescent females: Implications for assessment. 132;:ng ef Youth and

Adolescence. 13. 47-60.

Gilligan, C. (1982). In a different veig. Cambridge: Harvard University
Press.

Greenberg, M.T., Siegel, J .M. & Leitch, C.J. (1983) The nature and
importance of attachment relationships to parents and peers during
adolescence. leumalJLYmrthaadAdeleseenee. .1_2. 373-386.

Grollman, EA. (1987). W. Batesville Management Services.

Hansburg, H.G. (1972). W Vol. 1. Springfield,
IL.: Charles C. Thomas.

Hansburg. KG. (1980). WWW.
Huntington, N .Y.: Robert Krieger.

Hotch. D. (1980). WW

1 vi rn l s (Doctoral dissertation, Michigan State
University. 1979). DissertaticmAhstraeantemtienal. £1. 34378.

Irion, J.C., Coon, R.C., & Blanchard-Fields, F. (1988). The influence of

divorce on coping in adolescence. We; .11.
135-145.

 

104

Josselson, R. (1980). Ego development in adolescence. In J. Adelson (Ed.),
n b l n chol (pp. 188-210). New York: Wiley and

Sons.

Kandel, D. B. & Davies, M. (1982). Epidemiology of depressive mood in
adolescents. W, 39, 1205- 1212.

Kelly, J.E. & Wallerstein, J .S. (1976). The effects of parental divorce:
experiences of the child in early latency. Ameriean .19an at
LEMMA—til. £6. 20-32.

Kovacs, M. & Beck, A. (1977). An empirical-clinical approach toward a
definition of childhood depression. In 1.6. Schulterbrandt & A. Raskin
(Eds.), .‘01' 'r'n chem Dru i l‘:l‘_ at, turn

nrtxlels (pp. 1-25). New York: Raven.

Larson, R. & Richards, M.H. (1991). Daily companionship in late childhood
and early adolescence: Changing developemntal contexts. Child

MW 62. 284-300.

Laufer, M. (1966). Object loss and mourning during adolescence. In R. S.
Eissler. et al (EdS). WWW Vol 21 (pp

269- 293). New York. International Universities Press.

LaVoie, J. ( 1976). Ego identity formation in middle adolescence. W
lemharriAdeLescem. 5. 371-385.

Long, B.H., Ziller, R.C., & Henderson, EH. (1968). Developmental changes
in the self-concept during adolescence. The Scheel Review, ‘_7_6, 210-230.

Mahler, M.S., Pine, F., & Bergman, A. (1975). Th h 1 ° i f
hnman infant. New York: Basic Books.

Main, M., Kaplan, N ., & Cassidy, J. (1985). Security in infancy, childhood,
and adulthood: A move to the level of representation. In I. Bretherton
and E. Waters (Eds.), Growing points of attachment theory and

research. n h f i f R 'n hi1
Magnum, 5.9. (1.2 Serial No. 209).

Marcia, J .E. (1980). Identity in adolescence. In J. Adelson (Ed.), Handboek
W3! (pp. 159-187). New York: Wiley & Sons.

105

Marris, P. (1982). Attachment and society. In C.M. Parkes and J. Stevenson-
Hinde (Eds.), l f hm n ' m n eh vior (pp. 185-201).
New York: Basic Books.

Maultsby, M.C. (1980). r i e m ' n w 11- ein ' A han b ok of
mummmunseflng. Lexington. Rational Self Help Aids.

Mazor, A. & Enright, P. (1988). The development of the individuation

process from a social-cognitive perspective. Jenrnal ef Adelesceng, l_l_,
29-47.

McCombs, A. & Forehand, R. (1989). Adolescent school performance
following parental divorce: Are there family factors that can enhance
success? Adelesm, 251, 871-880.

Melear, J .D. (1973). Children’s conceptions of death. LQuLnaLQLGenede
mm, m, 359-360-

Miller, J .B. (1976). fIfewfld a new psyehelegy ef wemen. Boston: Beacon.

Montemayor, R. (1982). The relationship between parent-adolescent conflict
and the amount of time adolescents spend alone and with parents and

peers. Child Develenment, 33, 1512-1519.

Montemayor, R. & Brownlee, J .R. (1987). Fathers, mothers, and adolescents:
Gender-based differences in parental roles during adolescence. Jamal

eLYmrtlLandAdoleseeuse. 1.6. 281-291.

Moore, D. (1987). Parent-adolescent separation: The construction of adulthood

by late adolescents. WM, 23, 298-307.

Moore, D. & Hotch, DP. (1981). Late adolescents’ coneeptualizations of
borne-leaving- lemaLethuduniAdfleseneee. 1.0. 1-10.

Moore, D. & Hotch, DP. (1982). Parent-adolescent separation: The role of
Parental divorce. JoumalotloathjndAdsLlescm. ll. 115-119.

Moore, D. & Hotch, DP. (1983). The importance of different home leaving
strategies to late adolescents. Marlee, L8, 413-416.

106

Murphey, E. ,Silber, E. Coelho, G, Hamburg, D. & Greenburg, I. (1963).
Development of autonomy and parent-child interaction in late

adolescence. AmedeanieumaLeLQrLonsycmaer. 33. 643- 652

Newman, B. M. (1989). The changing nature of the parent-adolescent
relationship from early to late adolescence. Adelesceng, 34, 914-924.

Nie, N, Hull, C., Jenkins, J, Steinbrenner, K, & Brent, D. (1975).

StatrstrealfiaekageforiheSQeiaLSeenm. (2nd Ed) St Louis
McGraw-Hill.

Piaget, J. (1967). Six nsyehelogieal stndies. New York: Random House.

Podell, C. (1989). Adolescent mourning. The sudden death of a peer. Cliniefl
Seeiflodcflmal. 11, 64-78.

Proulx, J. & Koulack, D. (1987). The effect of parental divorce on parent-
adolescent separation. WW, ]_6, 473-480.

Puig-Antich, J ., Chambers, W., & Tabrizi, M.A. (1983). The clinical
assessment of current depressive episodes in children and adolescents:
Interviews with parents and children. In D.P. Cantwell & G.A. Carlson

(Eds.), Affeetiye diserders in ehildhegi and adeleseence; An Mate
(157-179). New York: Spectrum.

Radloff, L. S. (1977). The CES-D scale. A self-report depression scale for

research in the general population. AW,
1, 385-401.

Radloff, L.S. & Locke, 8.2. (1986). The community mental health
assessment survey and the CES- D scale. In M.M. Weisman, J .K.

Myers, & C. E. Rose (Eds. ), gem mnnity snfleys ef psyehiatrie
dnslders, 4 (pp. 177-189). New Brunswick: Rutgers University Press.

Roberts, R.E., Andrews, J.A., Lewinsohn, P.M., & Hops, H. (1990).
Assessment of depression in adolescents using the Center for

Epidemiologic Studies Depression Scale. W. 2,
122-128.

 

107

Roberts, R.E., Lewinsohn, P.M., & Seeley, J.R. (1991). Screening for
adolescent depression: A comparison of depression scales. Journal ef

the Amen’ean Aeademy ef Child and Adeleseent Psyehiag, 30, 58-66.
Rosenberg, M. (1965). Smjety_and_th_e_Ad_Q_1e§eent_Sel_f_Inrage. Princeton:

Princeton University Press.
Rosenberg, M. (1979). Ceneeiling_tne_Se1L New York: Basic Books.

Rutter, M. (1986a). Depressive feelings, cognitions and disorders: A
research postscript. In M. Rutter, C. Izard & P. Read (Eds.),

WQWImn n lini r iv
(pp. 491-519). New York: Guilford Press.

Rutter, M. (1986b). Developmental psychopathology of depression. Issues and
perspectives. In M. Rutter, C. Izard & P. Read (Eds. ), QED—9.5m

WWW (pp 3- 30)
New York: Guilford Press.

Sabatelli, R. & Mazor, A. (1985). Differentiation, individuation, and identity
formation: The integration of family system and individual

developmental perspectives. Adeieseenge, _2_Q, 619-633.
Selye, H. I1974). We; Philadelphia: Lippincott.

Sroufe, L. A. & Waters, E. (1977). Attachment as an organizational construct.
Wm. _8, 1184-1199.

Stein, M., Levy, M.T., & Glasberg, HM. (1974). Separations in black and
white suicide attempters. Arehives efi Senegal Esyehiag, 31, 815-821.

Stierlin. H (1974) W

W New York:
Quadrangle.

Strober, M. (1983). Clinical and biological perspectives on depressive
disorders in adolescents. In D.P. Cantwell & G.A. Carlson (Eds.),

ff ' ’ r i hil ' An (pp. 97-
103). New York: Spectrum.

108

Sullivan, K., & Sullivan, A. (1980). Adolescent-parent separation.
Develepmentai Esychelegy, _l_6, 93-99.

Sweetland, R. C, & Keyser, D. J. (Eds.). (1986). Tests; A eemprehensive

fr 1 r 1' ‘, m o 010 '10 .- ', nd 0 sins (2nd
ed.). Kansas City: Test Corporation of America.

Valente, S.M. & Sellers, J .R. (1986). Helping adolescent survivors of suicide.

In C.A. Corr & J.N. McNeil (Eds.), Adoleseengeflrddeatn (pp. 167-
182). New York: Springer.

Wallerstein, J .S. & Kelly, J.B. (1980). mm
W. New York: Basic Books.

Waterman, AS. (1982). Identity development from adolescence to adulthood:
An extension of theory and a review of research. Develenmental

Emhologx. 1.8. 341-358.

Weiss, RS. (1982). Attachment in adult life. In C.M. Parkes & J. Stevenson-

Hindc (Eds.). Ibeplaezeflattaclmentflmtmanbehayim (pp- 171-
184). New York: Basic Books.

Weissman, M., Sholomskas, D., Pottenger, M., Pursoff, B., & Locke, B.
(1977). Assessing depressive symptoms in five psychiatric populations:

A validation study. Ameriean Jenmai ef Enidemielegy, m, 203-213.

White, RW. (1966). WWW
persenaljty (2nd ed.). Chicago: Holt, Rinehart and Winston.

Wolfelt, AD. (1987). Resolution versus reconciliation: The importance of
semantics. Inanates, 11-13.

Wolfelt, AD. (1988). Reconciliation needs of the mourners: Reworking a
critical concept in caring for the bereaved. Ihanates, 7-10.

Yanchyshyn, G. W. & Robbins, D. R. (1983). The assessment of depression in
normal adolescents- a comparison study. CanadanMef
25mg. 23 522-526.

YouniSS. J. & Smoller. J. (1985). Adolesrenuelationuvimmethersiamers.
and friends. Chicago: University of Chicago Press.

"‘lllllllllllllllllE5