 

u.,v.V.-:.:.5.s.
x. urn;
Saw. .3 ‘ +
. . v 9. 9.532.?
... d. ‘1

v!

.5 ..
ﬁﬁﬁ;
_ a..-......\h‘. I. ~ .
‘ V . t ...,s.n..f.::...$ﬁ . :3 i .
n... . Rwandw...‘ n 79....33...» ..a,x....u....: . dish ‘2 35;.
.. I; a." ‘C. in.” .351. .wxm.

‘ F ‘

i'
:1

‘z... ‘.\

3i

0
g;
u

e ‘
.1... . .x
u-SQ. v 1

*n
«J

a ‘
Fri-H"
03L} 5
h" '

‘ ..
H5. .2.
5 .
u:

i.»

a
x

I"
‘g E;
.‘i

\ .
1‘4“IQP$. i“ ‘

u. six} . ..v .5.» .3635 .f ..
.. 3.... inbuc. ‘M‘v. ..
.. .ﬁhvx..m.ﬁz

. .. p but!
m. whoa 4“»
n1

r

a"... V. 1

:y3\l 3h»..-)

uvv‘t It A Kitty...
. 3:». an

...\.uu‘.. I

\N, ‘
Tut.“
hm"

P

.P $0911!-
Ia I

lilpﬁuin ...|.L. .o . .b‘
.._ .n.v :Vmu Ivcn‘u

 

. (53.2.1.1;
. .. A r. an , .

y

9‘ .1945. E
wnuwb.
r 1

.l.’.
. '39

I u 3?.”- .rl:
Pbrli 0.... . a
.

1..) .w

1‘53...’ 1.!- .

4.0.;:.

:33. .1: r

, )I v: .,
\fois. u: .x
v 1::

?n t)
){v’z .
v fnl
.

. ’
i..~' i.

7-:hﬁ9xm .2!»u;..::9.
3. .. 12!: ...||‘
I ’ﬁlil '5 u,-14li
b.1155

- 5...:
...|.. :

a-Ivglr

1v; qr ‘f
|iflv 71?.45-5 .

l- . I‘LL .Edﬁ

.1I..|I&

IIA .. “at. 13..
91.47:...
1

«hr I), 1.39?»

v A. - I .
.anNchA.
:13} a!!!

. 1:! . . Octanpaﬂ .1
ifl-kr.’". C—
.ﬂrvtﬁ r..tvth.1§r
n'i’f‘t~.,§“
. EiiuVrE ”Sun;
(Iranivd at 2
v . I
Hf?»

, t
:35:
w

{.Iagcﬁuw I v I
‘ Hagar? $1. «:1th
.. it; E 0.1.3.2341;
. éhbﬂﬂﬂﬂt 2v." nit-j
8.3“)"; {riffs
v'cu. ﬂ.

.-

it}...a.z$:..i¢.
.vri..qz.?i..t.
.15. i-

.5... .
hurﬁmfrzsz. Jr
:12. 1:93.311;
.acilsvvFSoés. if}:
$9.35. :11! if}:
flirt: 4. ti:

étkti;

. 1.29:512512-
$3....

nun-Pvt»!

)Vlv vfrll Ern3iirl4"

A 2. I}

{I‘ll .09.?

» O. II “v.75 g It"
A: ll? fill: Iv .Irwﬁﬂ'fibtlleé

’1 ) tic-.5“! y'!’

. Lilof. . ’5‘:

IvITOIIIIOI“;O gall; n... f

.10
.. Ib-Ivvi

V u‘sl‘4l\|v
Insular yr:

.3. 11.! I! {A 5...?!” fibroit
. ._. h rlh'v‘irttffl l(
‘ IIIIDA.

up. ;«-/w.lun p\.. (lillv

 

 

 

 

 

 

r...:1l)|"

 

                  

HI

l/lllllllll/llllll

             

3

This is to certify that the

dissertation entitled

Changing Explanatory Style in Children

presented by

Abigail Ellen Gleason

has been accepted towards fulﬁllment
ents for

Ph.D. dqnwhu Ps cholo y
lllll‘llll lllx;llll9

ofthe requirem

Major professor

DMe August 29, 19

96

MS‘U Ill an Afﬁrmuliw' A (lion/liq 1411/ Oppnrlunily [ml/Inn's”

 

 

LIBRARY

Michigan State
University

 

 

 

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

 

 

 

 

DATE DUE I DATE DUE DATE DUE
4. ml
1 2 0 4 0 1

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

1/98 mlClHCJDatoDuopﬁS—p.“

CHANGING EXPLANATORY STYLE IN MIDDLE-SCHOOL CHILDREN

By

Abigail Ellen Gleason

A DISSERTATION

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

DOCTOR OF PHILOSOPHY

Department of Psychology

1996

ABSTRACT

CHANGING EXPLANATORY STYLE IN MIDDLE-SCHOOL CHILDREN

By

Abigail Ellen Gleason

Fifty-one middle school students volunteered to participate in
an after-school program designed to increase self-esteem. Children
were assigned to one of three treatment groups: a) explanatory style
training, b) relaxation training, and c) no-contact.

The explanatory style and relaxation groups received five
hours (one hour weekly) of training to foster optimistic explanatory
style or to promote relaxation skills, whereas students in the no—
contact group received no training during this phase. All children
completed measures before groups began (pre-test), immediately
following training (post-test) and ﬁve weeks after the conclusion of
training (follow-up). The no-contact group received explanatory
style training after follow-up data collection, and completed
additional measures after the conclusion of their training.

At pre-test, depressed children obtained signiﬁcantly lower
scores than non-depressed children on measures of perceived
competence, self-efﬁcacy, and explanatory style. Many of these
differences persisted at post-test and follow-up. Depressed children
in all three groups experienced a reduction of depressive symptoms

at post-test [explanatory style group: F (1,7) = 8.36, p < .05; relaxation

 

training group: F ( 1,4) = 8.00, p < .05; no-contact group: F ( 1,7) =
14.82, p<.01].

Results from pretest to follow-up indicated an increase in
optimism for positive events among the explanatory style group
[F (1,17) = 6.28, p < .05] During this time, the no—contact group
experienced increasing pessimism in the face of negative events
[F(1,17) = 33.63, p < .001] and a decline in overall optimism
[F (1,17) = 15.36, p < .001]. These results suggest that explanatory
style training may have a positive impact on optimism among
children, and may serve as a protective factor against the
development of pessimism in the face of negative events.

There were no significant differences between treatment
groups in perceived competence, self-efficacy, or persistence at any
time. It is speculated that longer interventions may be necessary to
produce positive changes in these areas.

Suggestions are made for improving future interventions

among middle school children.

ACKNOWLEDGMENTS

I have been fortunate to have the help and support of many
individuals during the course of this project. Of course, I am
especially indebted to Dr. Dozier Thornton, who gave generously of
his time in order to meet with me frequently, especially when the
STARS program was just a vague idea in our heads. At every step of
the way, he offered support and guidance, and acted as a true
mentor.

Also, I greatly appreciate the interest of the other members of
my dissertation committee. Dr. Norman Abeles and Dr. Robert
Caldwell offered excellent advice on ways of promoting participation
among students at Summit. 1 am especially grateful to Dr. Ray
Frankmann, for his help with statistical issues, and his reminder to
have fun with the data analysis portion of the project.

Once the project started to take shape, Francine Edwards of the
Office of State and Federal Programs was instrumental in connecting
me with Summit Middle School and helping me to iron out some of
the details of the program. I am thankful to Chester Hughes, the
principal at Summit, who opened his school to me, and offered moral
support, advice, and his presence at a number of STARS-related
functions.

The wonderful staff at Summit also deserves special praise and
heartfelt thanks. Mrs. Leavitt, Mrs. Stinett, and Mrs. Loper were

always available and cheerful toward me, and helped make my work

iv

at Summit a true pleasure. I also wish to thank all the ﬁfth and sixth
grade teachers; they do an incredible job under challenging
circumstances. For their warmth and obvious dedication to the
students at Summit, I thank Mrs. Perry, Mrs. Brooks, Mrs. Martinez,
Ms. McZee and Mr. Sharp. A special thanks goes to Mrs. Peterson,
M.S.W., for her unending moral support, interest in the STARS
program, and commitment to the students.

I am especially appreciative of the students who participated
in the STARS program. Throughout our time together, I was
impressed by their intelligence, good humor, and patience. Bravo
also to their parents, who offered their feedback, and asked great
questions about the program.

In addition, I thank Amy Jadde Wong, who revived her role as
research assistant (from the days of my Master's thesis) to help me
in many phases of the data collection. Thanks also to Jamie Mogush
and Terry Hicks for their assistance in data collection.

My friends and family have been fantastic listeners throughout
my graduate school experience. I am so fortunate to have my family
as a built-in support system; thanks to Frank, Mary, Bridget, Ana,
Molly, Julia, and especially my mother. Thanks also to Stephanie
Gilbert, who has been a true inspiration to me, and has taught me a
lot about priorities.

1 am fortunate to have a number of wonderful long-distance
friends who have given me tremendous support and have helped
keep MCI in business. They include Rob Glass, Matt Jorgensen,
Sridhar Narayanan, "Fish" and Priya Ramakrishnan, Kristin Schard, C.

Sudhama, and Gloria Yueh.

My support system at MSU has also been extremely helpful to
me. Special thanks to larry Holmes, Beth Katz, Mark Heim, Stephen
Jefferson, and Alison Ward, for their encouragement and belief in my
graduation potential! I am also grateful to the support staff in the
psychology department. Cheryl Forcia and Suzy Pavick have given
me the beneﬁt of their knowledge of the program, and of life in
general, more times than I can remember.

I also want to express my appreciation to Natalie Denburg.
More than a wonderful friend for the last ﬁve years, she volunteered
to lead two STARS groups. I cannot adequately convey how much
her support has meant to me during my years at MSU.

Finally, I thank Rama Divakaruni, who has stuck with me for
the last six years. More than once, he has preserved my mental
health with his patience, kind words, and unconditional support. I

consider myself unbelievably lucky to have him in my life.

vi

TABLE OF CONTENTS

LIST OF TABLES ............................................. x
LIST OF FIGURES ............................................ xii
INTRODUCTION ............................................... 1
Childhood Depression .................................... 3
Learned Helplessness ................................... S
Explanatory Style ....................................... 7
Research with Adult Participants: Explanatory Style and
Depression ............................................ 8
Explanatory Style and Cognitive Performance ............. 1 O
Explanatory Style and Physical Health .................... 10
Research with Child and Adolescent Participants: Explanatory
Style and Depression .................................. I 1
Attributional Style and Academic Persistence ............. 1 3
Modiﬁcation of Explanatory Style ........................ 15
Use of Training or Therapy groups with Children ........... 1 8
Parental Involvement in Treatment/ Training of Children . . 22
Methodological Issues .................................. 23
Hypotheses ........................................... 24
Method ................................................... 26
Subjects .............................................. 26
Children's Test Battery .................................. 27

vii

Children's Depression Inventory .................... 27

Children's Attributional Style Questionnaire .......... 28
Perceived Competence Scale for Children ............ 28
Self-Efﬁcacy Scale ................................ 28
Adult Test Battery ..................................... 29
Demographics Questionnaire ....................... 29
Child Behavior Checklist ........................... 29
Parent Questionnaire ............................. 30
Procedure ............................................. 30
The Parent Meeting .................................... 3 1
The Explanatory Style Group ............................ 34
The Relaxation Training Group .......................... 36
The No—Contact Control Group ........................... 36
Post-test ............................................. 37
Follow-up ............................................ 37
Overview of Parent and Child Measures .................. 38
Results .................................................... 42
Persistent Differences Between Depressed and Non-Depressed
Students ...................................... ‘ ...... 42
Hypothesis 1 .......................................... 43
Hypothesis 2 .......................................... 50
Additional Findings Regarding the First TWO Hypotheses . . . SO
Hypothesis 3 .......................................... 57
Additional Findings Regarding Hypothesis 3 .............. 57
Hypothesis 4 .......................................... 61
Hypothesis 5 .......................................... 64
Additional Findings Regarding Hypothesis 5 ............... 66

viii

 

Hypothesis 6 .......................................... 68

Discussion ................................................. 70
Changes in Depression Status and Behavior Problems ...... 71
Explanatory Style ..................................... 74
The School Environment ............................... 74
Promoting Explanatory Style ............................ 76
Optimism for Positive and Negative Events ............... 77
Perceived Competence and Self-efﬁcacy .................. 78
Persistence ........................................... 79
Persistence and Depression ............................. 80
The No—contact Group .................................. 82
Future Directions ...................................... 83

List of References ........................................... 85

APPENDICES

APPENDIX A: Sources of Child Measures
APPENDIX B: Sources of Adult Measures
APPENDIX C: Homework Assignments
APPENDIX D: Anagrams

APPENDIX E: Parent Questionnaire

LIST OF TABLES

Table I - Sample Sizes of Depressed and Non-Depressed Children in

Each Treatment Group ............................ 32
Table 2 - Data Collection Procedures for All Groups ............. 39
Table 3 - Data Collection Schedule for Parent Measures ......... 41

Table 4 — Sample Sizes, Means, Standard Deviations (3, _M_, g) and t—
tests for differences between means for Depressed and
Non-Depressed Participants ........................ 44

Table 5 - Sample Sizes, Means, Standard Deviations (n, M1 32) and t-
tests for differences between means for Depressed and
Non-Depressed Participants ........................ 45

Table 6 - Analysis of Variance for Change in CD1 Scores Among
Depressed Children in the Explanatory Style Group . . . 47

Table 7 - CDI Scores for Depressed and Non-Depressed
Children in the Explanatory Style Group ............ 48

Table 8 - Problem Scores for Depressed Children in the
Explanatory Style Group .......................... 49

Table 9 - Analysis of Variance for Change in CD1 Scores Between
Pre-test and Post-test Among Depressed Children in the
Relaxation Training Group ......................... 5 1

Table 10 - CDI Scores for Depressed and Non-Depressed Children

in the Relaxation Training Group ................... 52

 

Table 1 1 - Problem Behavior Among Depressed Children in the
Relaxation Training Group ......................... S 3
Table 12 - CDI Scores for Depressed and Non-Depressed Students

in the No—Contact Group ........................... 55
Table 13 - Problem Scores for Depressed Children in the

No—Contact Group ................................ 56
Table 14 - Explanatory Style and Group Membership ........... 58

Table 15 - Correlations between CPCN Scores and Indices of
Perceived Competence ........................... 65

Table 16 — Cross-Out Persistence of Depressed and Non-Depressed
Participants ...................................... 67

Table 17 - Anagram Persistence of Depressed and Non-Depressed

Participants ...................................... 69

xi

LIST OF FIGURES

Figure 1 - Composite explanatory style (CPCN) by treatment
group at each data collection point .................. 60
Figure 2 - Explanatory style for negative events (CN) by treatment
group at each data collection point .................. 62
Figure 3 - Explanatory style for positive events (CP) by treatment

group at each data collection point ................. 63

xii

INTRODUCTION

Martin Seligman (1991) deﬁned explanatory style1 as the
manner in which one consistently explains the causes of events (p.
15). Seligman, along with a number of other investigators, has
proposed that certain types of attributions for negative life events
(i.e., internal, global, and stable attributions) can lead to depression,
as well as poor physical health and general inactivity or passivity
(Abramson, Seligman, & Teasdale, 1978; Peterson, Seligman, &
Vaillant, 1988; Seligman, 1991).

At this point, it may be helpful to consider what constitutes a
maladaptive attribution for a negative event. Internal attributions
tend to credit or blame oneself for events (e.g., "It was my fault that
our team didn't win" ). Global attributions generalize the success or
failure to other areas (e.g., "I'm not good at anything"). Stable
attributions contain the notion that situations will not change over
time (e.g., "I will never be any good at math"). The examples given
above illustrate the kinds of thoughts that may be related to the
development of depression in individuals with pessimistic
explanatory styles.

The concept of explanatory style derives from the reformulated
learned helplessness model of depression (Abramson et al., 1978).
The main idea of the original learned helplessness model is that the

expectation of uncontrollable outcomes leads to cognitive,

 

1 Throughout this paper, the terms "explanatory style" and "attributional
style" are used interchangeably.

motivational, and emotional deﬁcits. In the reformulated model, the
authors hypothesized that individuals who tend to make internal,
stable, and global attributions for negative life events are prone to
helpless behaviors and depression. The reformulated model posits
that these types of attributions lead to the perception of
helplessness, and to depression. In the same way, research in
explanatory style has identiﬁed internal, stable, and global
attributions for negative events as the components of maladaptive2
attributional styles (Seligman, 1991).

Although much of the learned helplessness literature
concentrates on depression as a consequence of maladaptive
attributions, studies in explanatory style suggest a number of
additional negative consequences associated with maladaptive
attributional style. For example, researchers conducting a thirty-ﬁve
year longitudinal study found that a pessimistic explanatory style
constituted a signiﬁcant risk factor for physical illness (Peterson,
Seligman, & Vaillant, 1988). In addition, ﬁndings in educational
research suggest a link between attributions and academic
persistence behaviors in children (Andrews & Debus, 1978; Chapin &
Dyck, 1976; Fowler & Peterson, 1981; Okolo, 1992).

Furthermore, there is evidence that explanatory style becomes
more stable as children age, and may begin a solidiﬁcation process
around puberty (Nolen—Hoeksema, Girgus, & Seligman, 1986; 1992).
If it is true that depressive explanatory style is related to depressed

 

2 The adjectives "maladaptive," "depressive," and "pessimistic" are used
interchangeably in describing attributional styles. All of these descriptors
refer to an attributional style characterized by internal, global, and stable
attributions for negative life events.

mood, poor physical health, and a lack of persistence, it makes sense
to target this explanatory style for intervention, in order to prevent
or reduce potential negative consequences and enhance physical and
mental well-being. In addition, it may be particularly useful to work
with children as they are entering puberty to determine whether
explanatory style training serves a preventive purpose.

This investigation sought to determine the efﬁcacy of a school—
based program directed at changing children's maladaptive
attributions. In light of evidence that optimistic explanatory style is
related to greater persistence behaviors as well as a reduction in
depressive symptomatology, it was expected that children who
participated in this program would evidence some positive changes
in both of these areas.

The following sections will describe the problem of childhood
depression, introduce the reformulated learned helplessness model
(Abramson, Seligman, & Teasdale, 1978), and explain the subsequent
development of explanatory style as a research topic. Later sections
will outline some attempts at modifying explanatory style, and will
provide examples of various cognitive treatment approaches that
have already been used with children and adolescents. Next, there
will be a discussion of parental involvement in interventions with
minors. Finally, a number of methodological issues related to this
type of research will be addressed.

Childhood Depression

There is evidence that feelings of depression increase between
childhood and adolescence (Rutter, 1986), especially for girls (Offord
et al., 1987). At present, there is a dearth of methodologically sound

 

epidemiological studies to determine the prevalence of depression in
childhood. Fleming and Offord (1990) conclude that rates of pre-
pubertal depression are probably less than 3%, and increase with
age. They temper their conclusion with the caveat that
methodological shortcomings and inconsistent use of assessment
instruments have probably led to under-reporting of depression
rates.

The Ontario Child Health Study (Boyle et al., 1987; Offord et al.,
1987) reported somewhat higher rates of emotional disturbance.
These investigators found that 8.3% of rural children and 10.7% of
urban children (ages 4-16) suffered from some type of emotional
disorder.3 In contrast to the aforementioned study by Fleming and
Offord (1990), these authors found that more than 10% of their
sample of children aged 41 I suffered from an emotional disorder.

Developmental issues appear to play a large role in childhood
depression. In pre-pubertal children, sex differences are apparently
negligible, with girls and boys showing similar rates of symptoms.
However, in adolescence, females surpass males in rates of
depression (Fleming & Offord, 1990). This pattern mirrors that of
adult depression. However, it is not clear at this time whether
females show more depressive explanatory styles than males at any
point in the lifespan.

Childhood depression may have consequences for academic
performance. For example, Nadine Kaslow and her associates

(Kaslow, Tannenbaum, Abramson, Peterson, & Seligman, 1983) found

 

3 " Emotional disorder" includes affective disorders, anxiety disorders, and
obsessive-compulsive disorder.

an inverse relationship between depression and problem-solving
performance.4 In their study, fourth and ﬁfth grade children
completed the Children's Depression Inventory (CDI), in addition to
block design and anagram tasks. CDI scores were inversely related
to problem-solving performance, indicating that depressed children
did not perform as well as non-depressed children in these cognitive
tasks.
Learned Helplessness

One model of depression which has been useful in cognitive-
behavioral therapy is the reformulated learned helplessness model
(Abramson et al., 1978). [earned helplessness is a process in which
an organism is exposed to an inescapable aversive stimulus and
subsequently learns that outcomes are not contingent upon
instrumental responses. In humans, this realization leads to four
major types of deﬁcits: cognitive, motivational, emotional, and self-
esteem. The cognitive deﬁcit in this model refers to the individual's
difﬁculty learning new information, speciﬁcally, learning that a
particular response will allow escape from the aversive stimulus.
The motivational deﬁcit involves retardation of voluntary
responding, such that the individual eventually stops trying to
escape the aversive situation. The emotional deﬁcit represents
feelings of depression. Finally, the self-esteem deﬁcit results from
the perception that although one is not capable of escaping an
aversive situation, others would be able to escape by responding

effectively.

 

4 Measures of cognitive performance included block designs (r = -.64) and
anagrams (r=-.67).

The reformulated model distinguishes between universal
helplessness and personal helplessness. Universal helplessness
refers to the belief that anyone would be helpless in a particular
situation, whereas personal helplessness refers to the belief that
oneself alone would be helpless in a given situation, whereas others
would not be helpless. The contribution of this distinction is that
personal helplessness involves an internal attribution, and thus is
more likely than universal helplessness to lead to depression and a
self-esteem deficit.

Another ﬂaw in the original model was an inability to predict
the chronicity and generality of helplessness. The authors of the
reformulated model posit that the perception of helplessness is
attributed to a cause which may be global or speciﬁc, stable or
unstable, and internal or external. If a global attribution is made for
a failure, the individual will likely generalize feelings of helplessness
to other situations, and perceive them as uncontrollable. In the same
way, if the individual makes a stable attribution for failure, the
perception of helplessness is likely to persist in the future.

The reformulation proposes that a subset of depression, called
helplessness depression, involves internal, global, and stable
attributions for failure, and external, speciﬁc, and unstable
attributions for success (Abramson, Seligman, & Teasdale, 1978).
The depressive attributional style characterized by internal, global,
stable attributions for negative events has also been referred to as
"pessimism" (Seligman, 1991).

According to the reformulated model, depression can result

from a chain of events which begins with a negative outcome and

ends with internal, stable, and global attributions or expectations of
hopelessness (Alloy, Abramson, Metalsky, & Hartlage, 1988). The
authors of the reformulated model (Abramson et al., 1978) proposed
a diathesis-stress relationship between outcomes and depression.
Whereas positive uncontrollable outcomes can lead to motivational
and cognitive deﬁcits, negative outcomes are necessary to produce
emotional deﬁcits, or depression. The authors do not, however, rule
out the possibility of other types of depression, with varying
etiologies.

It has been proposed that learned helplessness may be induced
through aversive stimulation that is inescapable (Hiroto & Seligman,
1975). Such a situation could easily occur in a school setting in which
a child is exposed repeatedly to failure and ﬁnally stops trying to
achieve. The end result of this scenario might be a lack of academic
persistence (motivational deﬁcit), a failure to recognize that studying
could improve performance (cognitive deﬁcit), depression (emotional
deficit), and a belief that others are capable of academic success
while this particular student is not (self-esteem deﬁcit).
Unfortunately, such a pattern could lend itself to further school
failure and corroborate the student's pessimistic attributions.
E_xplan_atorv Stvle

Philip Kendall (1993) hypothesized that cognitive dysfunction
takes on two different forms among children and adolescents.
Cognitive deﬁciencies are present when youths fail to thoughtfully
consider the consequences of their actions, and are associated with
impulsivity. On the other hand, cognitive distortions are present

when youths process information in a biased or dysfunctional

manner. Cognitive distortions are related to depression and anxiety
in youths. In both types of cognitive dysfunction, coping skills are
impaired.

Albert Ellis (1962) asserted that cognitive distortions
constituted both a cause and a maintenance factor in emotional
disturbance. He proposed that, if individuals could be released from
illogical thought patterns, they would probably not become
disturbed, and even if they did, they would not be able to remain
disturbed for very long. In the same way, optimistic explanatory
style does not imply an unrealistic worldview, but rather a realistic
openness to possibilities for success and positive outcomes, with the
understanding that a particular negative outcome is only one of
many possibilities. Optimistic explanatory style involves seeing
oneself as an agent rather than a passive onlooker. In other words,
optimists believe that they can inﬂuence outcomes by their actions.

It is important to note that an optimistic explanatory style is
not the equivalent of "blind optimism." That is, an optimist can be a
realist who does not expect the worst imaginable scenario to occur
when a situation has the potential to lead to disappointment. Rather,
an optimist would be likely to consider a number of possible
outcomes, and would probably generate and pursue a number of
alternative actions, should the outcome be disappointing. In this
way, the optimistic person would work to inﬂuence outcomes.
Research with Adult Participants: Explanatory Style and Depression

Research on explanatory style with adult subjects indicates a
positive correlation between attributional style and depression.

Using a sample of 39 adult patients with unipolar depression and 12

 

with bipolar depression, Martin Seligman and his associates
(Seligman et al., 1988), found that the worse the depression, the
more maladaptive the explanatory style. The patients in this study
received cognitive therapy, based on the program outlined by Beck,
Rush, Shaw, and Emery (1979). Some of the basic components of this
approach include translating the individual's concerns into solvable
problems5, helping the individual to confront and challenge irrational
beliefs and assumptions, and assigning "homework" to be done
between therapy sessions. Participants experienced both a reduction
of depressive symptoms and an improvement in explanatory style.
These changes were largely maintained at one year follow-up.

Similarly, another investigator found that depressed college
students were more likely to attribute failure to an internal cause,
and success to an external cause than their non-depressed
counterparts (Rizley, 1978). This pattern is consistent with the
reformulated learned helplessness model.

Research on cognitive correlates of depression has not yielded a
clear causal relationship between depression and attributional style.
In a one-year longitudinal study, investigators did not ﬁnd cognitive
measures predictive of future depression (Iewinsohn, Steinmetz,
Larson, & Franklin, 1981). These researchers concluded that
depressive cognitions coincide with other depressive symptoms, but

do not necessarily precede depressive episodes. However, they also

 

5 While a client might complain of "just not feeling right" upon entering
therapy, this program would encourage client and therapist to target
particular areas of the client's life for intervention. In this way, a client could
set speciﬁc goals such as "I‘d like to be more assertive at work" or "I'd like to be
more clear in communicating my feelings to my romantic partner." In so
doing, the individual may feel that his/her problems are manageable.

 

10

reported that individuals who showed more negative cognitions
during depressive episodes were slower to recover from depression
than subjects who displayed fewer depressive cognitions.
Explanatory Sgle and Cognitive Performance

Pessimistic explanatory style may also be related to poor
performance on cognitive tasks. There is evidence that anxious and
depressed adults experience difﬁculties with cognitive problem-
solving tasks and experience more cognitive interference than
control subjects when performing these tasks. In a study by
Zarontonello, Slaymaker, Johnson, and Petzel (1984), anxious and
depressed college students (as measured by the State Trait Anxiety
Inventory-A and the Beck Depression Inventory) performed more
poorly on a test of anagrams, thought more about how well they
were doing while completing the task, and evaluated their own
performance more negatively than control subjects.
Explanatory Style and Physical Health

Using data from a 35-year longitudinal study, Peterson,
Seligman, and Vaillant (1988) concluded that explanatory style has
important implications for physical health in adults. Investigators
from this research project followed 99 men from Harvard
University's classes of 1942-1944. The participants were screened
so that only the individuals who were believed to be in very good
mental and physical health, with excellent academic records, were
selected for participation. Between ages 25 and 40, there was very

little correlation between health at the present age and explanatory

11

style at age 25.6 However, at age 45, the correlation jumped
dramatically (r = .37, p < .001 ), indicating that explanatory style at
age 25 was predictive of future physical health. The authors
suggested a number of possibilities for the relationship, namely that
the pessimistic subjects may have been more passive than their
peers in the face of illness. In other words, individuals with a
pessimistic attributional style might have failed to seek or follow
medical advice, or might have neglected to attend to basic preventive
health care measures.

Peterson and Bossio (1991) posit that a combination of blaming
oneself for bad events and feeling powerless to change one's
situation creates a great deal of stress, which weakens the immune
system over time, and makes illness more likely.

There is a growing body of evidence that pessimistic
explanatory style is related to, but not predictive of, depression,
cognitive performance deﬁcits, and poor physical health in adults.
The next few sections will contain descriptions of research related to
explanatory style in children and adolescents.

Research with Child angLAdolescent Particinmgplanimyityle
and Depression

Research on explanatory style among children and adolescents
has also offered some support for the reformulated learned
helplessness model. In other words, the literature suggests that
depressed children have more pessimistic attributional styles than

non-depressed children (Garber, Weiss, & Shanley, 1993; Kaslow,

 

6 Controlling for initial physical and mental health, the correlations ranged
from .03 to .13. None of these correlations was significant at the .05 level.
Measures of physical health were taken every ﬁve years.

12

Rehm, & Siegel, 1984; McCauley, Mitchell, Burke, & Moss, 1988;
Nolen-Hoeksema, Girgus, & Seligman, 1986; Nolen-Hoeksema, Girgus,
& Seligman, 1992). Further, there is evidence that the correlation
between depression and explanatory style increases with age,
lending support to the notion that explanatory style becomes
increasingly stable with age (Nolen-Hoeksema et al., 1992).

Although research has tended to focus on children's
attributions for negative or stressful life events, there is evidence
that depressed children also View positive life events differently
than non-depressed children. In a study by Elizabeth McCauley and
her associates (McCauley, Mitchell, Burke, & Moss, 1988), it was
found that depressed children made more external, unstable, and
speciﬁc attributions for successes. This pattern of dismissing
successes as external to oneself and one's abilities may prevent the
individual from engaging in proactive behaviors, and thus maintains
the perception of helplessness.

Like Lewinsohn and his colleagues (1981), Susan Nolen-
Hoeksema and her associates (1986) found conﬂicting results about
the direction of causality between explanatory style and depression.
In a one-year longitudinal study, they reported that explanatory
style was predictive of future depression in children, and that
depression also predicted changes in explanatory style. Five times
during the course of a year, the researchers administered the
Children's Attributional Style Questionnaire (CASQ; Seligman et al.,
1984) and the Children's Depression Inventory (CD1; Kovacs, 1980),
among other instruments, to a group of third, fourth and fifth grade

students. They found that explanatory style at time n served as a

13

predictor of depression at time n + 1, even when depression at times
12 and n — 1 were controlled. The investigators conducted a similar
analysis to determine whether depression scores predicted changes
in explanatory style, and found that depression at time n was
predictive of explanatory style at time n + 1 . This pattern of results
suggests that depression and explanatory style coincide and may
reinforce one another. For example, the authors posited that
depression could play a role in the development of a pessimistic
attributional style, and that a maladaptive attributional style could
also contribute to depression. Taken together with the ﬁndings of
Lewinsohn and his associates (1981), it appears that, at the very
least, pessimistic explanatory style may be related to a delay in
remission from depressive symptomatology.

Attributional Style and Academic Persistence

Much of the educational literature in explanatory style has
focused on its relationship to persistence. One general ﬁnding has
been that attributions have an impact on persistence behaviors in
children (Andrews & Debus, 1978; Chapin & Dyck, 197 6).
Investigators in this area have attempted to encourage persistence in
children by teaching a relationship between effort and outcomes.

In a study by Andrews and Debus (1978), sixth graders who
attributed failures to lack of effort (internal, unstable) tended to be
more persistent than children who attributed negative outcomes to
lack of ability (internal, stable) or task difﬁculty (external, unstable).
Furthermore, the investigators found that children who were
verbally reinforced for associating effort with outcomes tended to be

more persistent at post-test, and that there was generalization of this

14

persistence to an unfamiliar tester and tasks that were not used in
the original testing sessions.

Fowler and Peterson (1981) expanded on these ﬁndings and
conducted attribution retraining sessions with middle school subjects.
In these training sessions, an experimenter either verbally associated
outcomes with children's level of effort or taught the children to say
to themselves, "I got that right. That means I tried hard." or "No, I
didn't get that. That means I have to try harder." Children in the
control groups were merely given feedback on whether their
responses were correct or incorrect. Children who received
attribution retraining were more persistent in reading difﬁcult
materials than control subjects on later trials. It is important to note
that these children persisted in the face of failure, suggesting that
attribution retraining may reduce some of the negative effects of
school failure. This type of intervention may allow children to
continue to learn rather than abandon tasks as they become more
difﬁcult.

In a more recent study (Okolo, 1992), children with learning
disabilities participated in eight 30-minute sessions involving a
computerized mathematics program. Members of the experimental
group received feedback about the number of correct responses they
had given, and whether or not they had reached their goals.7
Experimental participants also received some attributional feedback,
relating their performance to effort and ability if they succeeded,
and to effort alone if they failed. On the other hand, members of the

 

7 For all participants, the speciﬁed goal was to solve the same set of problems
three times within a given time period, and achieve at least 80% accuracy.

15

control group only received feedback on the number of correct
responses and the status of the goal. Although the students did not
differ on the types of attributions made for success and failure at
post-test, those who received attribution retraining voluntarily
completed more problem sets than the control subjects. In addition,
despite the similarity between the two groups on a computation pre-
test, children in the attribution retraining group performed
signiﬁcantly better on a computation post-test than the children who
received neutral feedback. These ﬁndings suggest that attribution
retraining has an impact on academic persistence. Improved
performance on the computational measure may have been a
reﬂection of persistence as well as the additional practice undertaken
by children in the experimental group.

The research literature on explanatory style in children and
adolescents resembles the adult literature in that both link
explanatory style to depression and cognitive performance. The key
difference between adults and children is that there is no evidence at
present that children with pessimistic explanatory styles suffer from
poor health, as compared with optimistic children. However, the
results of the study by Peterson and his colleagues (1988) suggest
that the correlation between explanatory style at age 25 and physical
health increases with age, and becomes pronounced only at around
age 45.

WW

According to Beck and Weishaar (1989), " the goals of cognitive

therapy are to correct faulty information processing and to modify

dysfunctional beliefs and assumptions that maintain maladaptive

16

behaviors and emotions." (p. 28) Certainly, pessimistic attributions
are often both irrational and self-defeating. For example, imagine
that a child experiences a failure, and then says, "I never could do
anything right." This statement is surely inaccurate, and the child
could probably come up with a memory of doing something kind or
helpful or intelligent. More importantly, however, this internal,
stable, and global attribution gives the child no reason to persist or to
believe that the future will be different. In that way, the child's
attribution maintains both maladaptive behaviors and negative
emotions.

On the other hand, consider a situation in which the same child
experiences a failure and says, "I guess I'll have to study harder to
improve my grade. This teacher tends to give pretty tough exams."
The attribution is now external (the teacher gives difﬁcult tests),
unstable (there is the expectation that the grade will change through
study), and speciﬁc (limited to this particular course). In this case,
the child may be encouraged to take action in order to improve the
situation, and will probably not experience motivational, cognitive,
emotional, or self-esteem deﬁcits in relation to the attribution.

There is evidence that cognitive therapy can help adults
develop a more optimistic explanatory style. For instance, in a study
by Martin Seligman and his colleagues (1988), depressed adults
participated in psychotherapy at the Center for Cognitive Therapy at
the University of Pennsylvania, and were monitored for depression
levels and explanatory style scores. The therapy was based on the
model outlined by Aaron Beck and his colleagues for the treatment of
depression (Beck, Rush, Shaw, & Emery, 1979). In this study, the

17

most severely depressed patients tended to have the most
pessimistic explanatory styles. The authors found an inverse
relationship between depression and explanatory style such that
decreases in depression were associated with more optimistic
explanatory styles. Furthermore, patients with optimistic
explanatory styles continued to be less depressed at one year follow-
up. These ﬁndings support the notion that explanatory style is
related to depression, and that cognitive therapy may facilitate
change in explanatory style.

Although some studies have examined the relationship
between attributions and performance in children, no studies have
focused solely on changing the explanatory style of children. There
is evidence that cognitive work and skill building can lead to a
reduction of depressive symptoms (J aycox, Reivich, Gillham, &
Seligman, 1994; Kahn, Kehle, Jenson, & Clark, 1990) or a reduction of
perceived severity of stressors (Elias, Gara, Ubriaco, Rothbaum,
Clabby, & Schuyler, 1986).

For children, in contrast to adults, changes in depression level
have not coincided with changes in explanatory style. In a study by
Lisa Jaycox and her colleagues (1994), a group of children at risk for
depression8 was given cognitive and social problem solving skill
training. Although depressive symptoms decreased among these
children, explanatory style did not change signiﬁcantly. However,

experimental subjects experienced a greater reduction of depressive

 

3 The high risk group was composed of children who showed depressive
symptoms as measured by the CDI, and who evidenced a perception of marital
conﬂict, as measured by the Child's Perception Questionnaire (Emery &
O'Leary, 1982).

18

symptoms than control subjects, and this difference was maintained
at six month follow-up. The control group in this experiment was
made up of a combined waitlist group and no-contact control group.
Thus, the role of non-speciﬁc factors in this study is not clear.

There are several possibilities as to why these investigators did
not find signiﬁcant changes in explanatory style. The most obvious,
of course, is that the children's explanatory styles did not change,
and thus no change could be measured. A second possibility, which
has been suggested previously, is that children's explanatory styles
grow more stable with age, and may not be consistent enough for
meaningful pre-/post- comparisons before puberty (Cole & Turner,
1993; Nolen-Hoeksema, Girgus, & Seligman, 1992). A third
alternative is that the current measures of explanatory style for
children are inadequate. Cole and Turner (1993) stated that the
Children's Attributional Style Questionnaire (CASQ) has poor internal
consistency, which may be related to children's lack of a consistent
attributional style. Finally, another possibility is that the
relationship between explanatory style and depression differs
between adults and children.
Use of Training or Therapy Groups with Children

A number of efforts have been made to reduce depression
levels in children and adolescents, or to reduce the negative impact
of common stressors on this population. Group treatments have
received more research attention than individual interventions, and
have shown promise in reducing depression (Elias, Gara, Ubriaco,
Rothbaum, Clabby, & Schuyler, 1986; Fine, Forth, Gilbert, & Haley,

19

1991; Jaycox, Reivich, Gillham, & Seligman, 1994; Lewinsohn &
Clarke, 1984).

Marc Zimmerman (1990) spoke of "learned hopefulness," which
he described as "...the process of learning and utilizing problem-
solving skills and the achievement of perceived or actual control." (p.
72) He proposed that the perception of present and future
controllability would lead to "hopeful" behaviors and emotional
states, including psychological empowerment, proactive behavior,
and reduced alienation. His ﬁndings indicated that a sense of
empowerment played a large role in reducing feelings of alienation.

A recent study compared cognitive-behavioral, relaxation, and
self-modeling treatments for depression in adolescents (Kahn, Kehle,
J enson, & Clark, 1990). The cognitive—behavioral and relaxation
training groups participated in twelve 50-minute sessions over a six-
week to eight-week period. The self-modeling condition, however,
involved twelve sessions, but only for 10—12 minutes each. In the
self-modeling condition, participants met with the experimenter
individually, rather than in a small group. For the cognitive-
behavioral group, an adolescent form of the Coping with Depression
(CWD) course (Lewinsohn & Clarke, 1984) was used. In each session,
a speciﬁc depression-related topic was addressed. The course was
based on a skill deﬁcit/ skill acquisition model in which participants
were encouraged to practice various skills during and between
sessions. The relaxation group also involved practice within and
between sessions, and included lecture, demonstration, discussion,
case examples, and feedback. The self-modeling participants were

instructed to behave in certain ways that were not consistent with

20

their depression (e.g., eye contact, erect body posture, positive
affective expression, positive verbal attributions). Subjects in all
treatment groups experienced a reduction in depressive symptoms.
However, subjects in the self-modeling condition did not generally
maintain the changes at one month follow-up, in contrast to
members of the cognitive-behavioral and relaxation training groups.
Another group of investigators (Butler, Miezitis, Friedman, &
Cole, 1980) conducted a school-based intervention with two
experimental groups (a role play group and a cognitive restructuring
group) and two control groups (a placebo control group and a no-
contact control group). Both experimental groups and the placebo
control group participated in ten weekly hour-long sessions. Each
role play session focused on a topic which would be played out by
group members and discussed. The topics were related to
depression, and included themes such as success and failure, peer
acceptance and rejection, and loneliness. The cognitive restructuring
group sought to teach children to replace dysfunctional automatic
thoughts with more constructive ones, to enhance listening skills, and
to understand the relationship between thoughts and feelings.
Children in the placebo control group were taught to solve problems
collectively by sharing research and information. The ﬁnal group
was a no-contact control group. The investigators found that
children who received role play training tended to experience
decreased depression levels, whereas children in the cognitive
restructuring and control groups did not differ significantly from one
another after the treatment. The authors speculated that the role

play group had more appeal for the participants, as it did not

21

demand as much verbal activity or introspection from the
participants.

One program which targeted middle school children took a
social problem-solving approach, and included both instructional and
application phases over an academic year (Elias et al., 1986). The
instructional phase was conducted throughout the ﬁrst semester of
an academic year, and consisted of an introduction to problem-
solving over 20 40-minute sessions. The application phase was
implemented throughout the second semester, and consisted of two
major components: formal sessions in which the teacher
demonstrated methods of integrating problem-solving skills into
regular class sessions and informal sessions in which the teacher
mediated conﬂicts by encouraging the children to use the problem-
solving skills they had learned. While the formal sessions of the
application phase were held approximately once a week, the informal
sessions were held at the teachers' discretion. The authors report
that these informal sessions took place approximately three times a
week.9 Children who received both phases of the training perceived
life stressors as less of a problem than children who received only
the instructional portion of the training. Certainly, part of the
success of this program may be related to its length and integration
into the academic curriculum.

The Penn Prevention Program included both cognitive and

social problem-solving components in an intervention designed to

 

9 The authors did not report individual differences across teachers, but merely
stated that, "...our data indicate that such interventions were used
approximately three times per week by the majority of teachers." (p. 265)

22

reduce or prevent depression among at-risk10 10—13 year olds
(Jaycox, Reivich, Gillham, & Seligman, 1994). Children who received
this treatment were signiﬁcantly less depressed than children in a
combined waitlist/no—contact group at the conclusion of the
treatment period, as well as at 6-month follow-up. Although these
children did not show corresponding changes in a composite measure
of explanatory style“, they were less likely to attribute events to
stable causes at the end of treatment.
Parental Involvement in Treatment/Training of Children

Although there have been many calls for greater parental
involvement, either as collaborators in treatment or as clients in
family interventions, little research has been done to determine the
efﬁcacy of such applications (Kendall, 1993). Iauren Braswell (1991)
cites three major reasons to involve parents in cognitive behavioral
treatment for their children. First, parents often play a role in
deﬁning the child's problem. Parents frequently refer their children
for treatment, and may possess important information about their
children's problems. Secondly, parents may have an effect on the
child's problem. For example, they may cause, exacerbate, or
otherwise moderate the problem. It may be helpful in treatment to
determine the impact of the parents' behaviors and communications
on the child's problems, and if necessary, work with the parents to
help them change maladaptive patterns. Finally, parents can be

 

10 Children were classified as "at risk" on the basis of elevated depression
scores (as measured by the Children's Depression Inventory; Kovacs, 1980) and
self-reports of parental conﬂict (as measured by the Child's Perception
Questionnaire; Emery & O'Leary, 1982).

1 1 Explanatory style was measured by the Children's Attributional Style
Questionnaire (CASQ; Seligman, Peterson, Tannenbaum, Alloy, & Abramson,
1984).

23

important collaborators in the therapeutic process, and may facilitate
skill transfer, or maintenance of therapeutic change.

Of course, it is often the case that parents resist participation in
treatment. Robin and Foster (1989) point out that a parent may not
see the need to participate in his or her child's treatment, since the
child or the other parent is perceived as "the problem." (p. 212)
They suggest contracting with families in order to clarify all
participants' expectations for treatment. They also emphasize the
importance of building rapport with family members, so that respect
is conveyed to all family members, and it is clear that therapy will
not be a forum for blaming or scapegoating any individual family
member. Ellen Wachtel (1994) asserts that resistance is a response
to anxiety, and thus the therapist should work to make clients feel at
ease, so that therapy can proceed. It seems reasonable that
experimenters would also beneﬁt from these guidelines, as
cooperation may be enhanced and attrition reduced if they make
efforts to respectquy involve family members.

Methodological Issues

A number of methodological issues have arisen in the study of
depression in childhood and adolescence. According to Fleming and
Offord (1990), one major stumbling block in this type of research is
sampling. Perhaps most importantly, samples are often too small to
permit conclusions about epidemiology of disorders. In addition, the
attrition rate is frequently so great as to compromise the
representativeness of the sample. Unfortunately, the subjects who

drop out of these studies are not usually followed up, and it is

24

unclear whether they differ demographically from the remainder of
the sample.

Another important problem in depression research with this
population is the inconsistent use of measures. Measures for
depression in children have not withstood as much psychometric
scrutiny as have adult measures (Kazdin, 1981). That is, issues of
reliability and validity do not always receive adequate attention,
which may result in the use of inappropriate measures.
Hypptheses

Martin Seligman (1991) proposed a method of teaching
children to become more optimistic, in order to reduce the risk of
depression, and allow children to persist even when faced with
disappointments and failures. Unfortunately, there is no research
known to the present author that has closely followed Seligman's
training approach, and it is unclear at this time whether a group
focused solely on changing the explanatory style of children would
be effective and beneﬁcial to participants.

If it is true that depressive explanatory style is related to
depressed mood, reduced persistence, and poor physical health (at
least in adulthood), it makes sense to target this explanatory style for
intervention, in order to prevent or reduce potential negative
consequences and enhance physical and mental well-being.

I hypothesize that (1) Depressed children who receive a ﬁve-
session training course (over five weeks) aimed at improving
children's explanatory styles will show a reduction in depressive
symptoms (as measured by the Children's Depression Inventory;

Kovacs, 1980) and a decrease in behavioral problems at home (as

25

measured by the Child Behavior Checklist behavior problems
subscale completed by parents; Achenbach & Edelbrock, 1983) at the
conclusion of the course.

(2) Depressed children in a control group (relaxation training)
may, also experience a reduction in depressive symptoms and
behavioral problems.

(3) Participants in an explanatory style group are expected to
show more optimistic explanatory styles at the conclusion of the
course than children in the relaxation training group and children in
the no-contact control group.

(4) Participants in the explanatory style group are also
expected to show greater perceived competence and conﬁdence in
their own skills and abilities (as measured by the Perceived
Competence Scale for Children; Harter, 1982; Child Behavior Checklist,
social competence subscale, Achenbach & Edelbrock, 1983) and a
greater sense of self-efﬁcacy (as measured by the Self-Efﬁcacy Scale;
Sherer, Maddux, Mercandante, Prentice-Dunn, Jacobs, & Rogers,
1982) as compared to participants in the other two groups.

(5) Members of the experimental group are expected to show
more persistence behaviors at post-test and follow-up than members
of the relaxation training group and the no-contact control group.

(6) After receiving explanatory style training, children in the
no-contact control group will show decreases in their depression
scores, and will show greater perceived competence, self-efﬁcacy,
and a more optimistic attributional style, as compared with members

of the relaxation training group.

26

Method

Subjects

The parents of all ﬁfth and sixth graders at a Michigan middle
school were asked to consent to a self-esteem screening of their
children. Children with elevated scores on the CD1 (CD1 2 14) and
children with low scores on the CD1 (CD1 _<5) were recruited to
participate in the study.12 The recruitment process consisted of
sending a letter home to parents of selected children and requesting
consent for these children to participate in an after-school program,
called the S.T.A.R.S. (Students Trying and Reaching Success) program.
Children who completed the program were promised S.T.A.R.S.
program t-shirts and an awards banquet at the end of the program.

Twenty-two boys and twenty-nine girls between the ages of 10
and 12 were included in the data analysis. According to parent
reports, 32 (63%) of the participants were African-American, 10
Caucasian (20%), 1 Hispanic (2%), 1 Native American (2%), 4 Multi-
racial (8%), and 2 were identiﬁed as "Other" (4%). One parent
declined to identify the ethnicity of her child (2%). Compared with
the population of the surrounding community, Caucasian students
were slightly under-represented in the program (Beecher
Community School District, 1 990).

 

12The parents of a child whose CDI score was 43 were notiﬁed that their child
showed some risk for depression. Information regarding the community
mental health system in the area and mental health resources available at the
middle school was provided. The parents were invited to call the investigator
for additional information or referrals.

 

 

27

Twenty-three (45%) of the parents reported a household
income of less than $10,000 a year; nine (18%) reported an income
between $10,000 and $20,000; ten (20%) reported $20,000 to
$35,000; four (8%) reported $35,000 to $50,000 and two (4%)
reported $50,000 to $70,000. Three parents (6%) declined to specify
their household incomes. No data were available regarding the
average income in this community.

The parents of all participants agreed to attend two meetings:
one before the training began, and one after the conclusion of
training. Parents who could not attend these meetings were asked to
come to the school to complete parent forms. Parents who were
unable to arrange transportation to the school were allowed to
answer questions over the telephone.

Parent and child measures were chosen for their wide use
among researchers as well as their high levels of reliability. It was
thought that the use of these particular materials would allow other
researchers to easily compare this study with related investigations.
Children's Test Battegﬁ13

Children's Depression Inventog (Q1; Kovacs, 1980). Each CDI
item contains three sentences, indicating different symptom severity
levels. Sentences are assigned a value from 0 to 2, with a score of 2
being the most severe level of the symptom. The total score ranges
from 0 to S4. The internal consistency of the CDI ranges from .80 to
.94 (Saylor, Finch, Spirito, & Bennett, 1984), and this measure has

been shown to reliably discriminate psychiatric patients from non-

 

13Sources of all child measures may be found in Appendix A.

 

 

 

28

referred adolescents (Carey, Faulstich, Gresham, Ruggiero, & Enyart,
1987).

Children's Attributional Style Questionnaire (& Seligman et
al., 1984). The CASQ is a 48-item forced choice measure in which
two dimensions of explanatory style are held constant, while the
third varies. In this way, children may receive composite
explanatory style scores for positive and negative events, in addition
to an overall composite score. The coefﬁcient alphas (Cronbach,

195 1) for the Composite Positive (CP), Composite Negative (CN), and
Total Composite (CPCN) are .71, .66, and .73, respectively.

Perceived Competence Scale for Children (Harter, 1982; Harter,
1987). Harter's Perceived Competence Scale for Children is a 36-item
scale in which children are asked to rate their own competence, in
comparison with other children, in ﬁve domains appropriate to
children: 1) scholastic competence, 2) social acceptance, 3) athletic
competence, 4) physical appearance, and 5) behavioral conduct. Each
item consists of two opposing statements, taking the form, "Some
kids..." but "other kids..." Children are asked to identify the
statement that is most like them, and then rate whether that
statement is "really true for me" or " sort of true for me." Items are
scored on a scale of 1 to 4, with 1 indicating low perceived
competence, and 4 indicating high perceived competence. The
coefﬁcient alphas for the cognitive, social, physical and self—worth
subscales range from .75 to .83, .75 to .84, .77 to .86, and .73 to .82,
respectively.

Self-Efficacy Scale (Sherer, Maddux, Mercandante, Prentice-
Dunn, Jacobs, & Rogers, 1982; Sherer & Adams, 1983). Self-efﬁcacy

29

involves optimism (the perception that one will be able to perform a
given task successfully) and a positive internal attribution (the
perception of oneself as a skilled or skillful individual). The Self-
Efﬁcacy scale contains 23 items pertaining to one's sense of general
self—efﬁcacy or one's sense of social self-efﬁcacy. The remaining 7
items are "ﬁllers" that relate to the subject's interests and hobbies.
Subjects are asked to rate each item on a 5-point scale as to whether
they "disagree strongly," "disagree moderately," "neither agree nor
disagree," "agree moderately," or "agree strongly" with each
statement. The coefﬁcient alphas for the General Self-efﬁcacy and
Social Self-efﬁcacy subscales are .82 and .71, respectively.
Adult Test Battery14

Demographics Questionnaire. Parents were asked several
questions about their age, educational level, marital status and socio-
economic status. They were also asked about their children's age,
grade level, and school performance.

Child Behavior Checklist (CBCL; Achenbach & Edelbrock, 1983).
The CBCL consists of 20 items assessing social competence and 1 18
items assessing behavioral problems. For the social competence
subscale, parents report on their children's participation and
performance in sports, hobbies, social activities, academic work, and
home behavior. For the behavior problems subscale, parents rate
their children on a number of behaviors. A rating of 0 indicates that
a statement is not true of the child, a rating of 1 indicates it is
sometimes true, and a rating of 2 indicates that it is often true of the

child. Test-retest coefﬁcients for the CBCL range from .95 (behavior

 

1"'Sources of all adult measures may be found in Appendix B.

30

problems) to .99 (social competence) after one week, and from .84
(behavior problems) to .97 (social competence) for a three-month
interval.

Parent Questionnaire. Parents were asked about a number of
activities in which their children participate (e.g., studying, playing a
musical instrument, or participating in a sport). For each activity,
parents were asked how long their children practice at one time, how
much they practice during the course of a week, and whether or not
they practice every day. Parents were also asked how often their
child "gave up" an activity, whether they were able to "redirect" their
child to continue to practice, and how long their child has been
participating in each particular activity.

Procedure

All ﬁfth and sixth graders who received parental permission
completed the CDI, the CASQ the Perceived Competence Scale for
Children, and the Self-Efﬁcacy Scale at an initial depression screening
session. Based on CDI scores, 30 depressed children and 40 non-
depressed children were recruited to participate in this study.

At the screening session, children were given a series of ten
ﬁve—letter anagrams to solve. All words that were included in the
set of anagrams were at or below a third grade reading level. The
ﬁrst set of ﬁve anagrams was solvable, while two of the second set of
ﬁve were unsolvable. Children were asked to do their best on this
task, and told that they could take as much time as they wanted.

The number of correct answers and the amount of time each child
spent on this task was recorded. At this time, they also completed a

test of persistence, which consisted of a number of cross-out tasks.

31

For each item, the child was asked to draw a line through the three
visual stimuli that resembled the target stimulus. The total number
of items completed became the persistence score.

Three types of groups were constructed, such that
approximately 10 depressed children and 10 non-depressed children
were invited to participate in each type of group. Efforts were made
to construct groups with similar male to female ratios. The group
leader was naive as to the status of group participants on the
screening measures. Table 1 provides sample sizes of depressed and

non-depressed children for each treatment group.

 

Insert Table 1 About Here

 

The experimental group was the explanatory style group in
which children were instructed in Seligman's (1991) method for
explanatory style change. One control group was a relaxation
training group in which children practiced relaxation techniques
aimed at reducing stress. The second control group was a no-contact
control group in which children had no contact with the
experimenter, except to complete pre-test and post-test measures.
There were two explanatory style groups and two relaxation training
groups, made up of 10 children each, so that each participant
received some individual attention from the group leader.

The Parent Meeting
Before the training began with the children, parents were

asked to attend a meeting in which the rationale of the project was

32

Table 1

Sample Sizes of Depressed and Non-Depressed Children in Each
Treatment Group

 

 

 

Group Type
Time Explan. Style Relaxation No—Contact
Pre-test (9,9) (5,8) (9,1 1)
Post-test (9,9) (5,8) (8,1 1)
Follow-up (9,9) (4,7) (8,10)
Post-treatment --- --- (4,5)
Tx Follow-up --- --- (3,5)

 

Note. n = (number of depressed children, number of non-depressed
children).

33

explained, and parental cooperation was solicited. If the child had
two parents, both were invited to attend the meeting and complete
the necessary forms. In order for a child to participate, at least one
parent was required to attend the meeting, or otherwise complete
the parent instruments.15

Parents were informed that two different types of groups
would be formed, and that the groups were designed to help children
to cope better with stressors. This investigation examines the
effectiveness of these groups.

Children were expected to complete "homework assignments"
between all group sessions.16 Assignments were checked (for effort,
not "correctness") at each group meeting. Parents were asked to
encourage their children to complete all of the assignments, so that
the children could derive maximum beneﬁt from participation and
the groups could fairly be compared. The parents of participants
were told that all of the children would eventually be able to
participate in a group, but some would have to wait several weeks to
be placed in a group.

All children who attended at least four of the ﬁve group
sessions were invited to attend a banquet, along with their parents,
at the conclusion of the program. Children were given S.T.A.R.S.
program t-shirts at that time. Data from children who attended less

than four sessions were not included in the ﬁnal analysis.

 

15 Parents were offered two dates for the parent meeting. If they could not
attend either meeting, the experimenter made efforts to make alternative
arrangements with them (e.g., telephone interviews, or appointments with
them in their homes or at the school).

16A description of homework assignments is provided in Appendix C.

34

At the end of the meeting, parents were asked to sign an
informed consent agreement and complete a brief demographics
questionnaire, the Child Behavior Checklist (CBCL; Achenbach &
Edelbrock, 1983) and a short questionnaire about their children's
levels of persistence at various tasks.

Notices were sent home to remind parents of the follow-up
parent meeting, scheduled approximately one month after the ﬁnal
children's group session. Parents were asked to attend the follow-up
meeting regardless of whether their children had been placed in a
group. Because of the small number of parents attending the ﬁrst
two meetings, and the small number of participants in the no-contact
group completing the explanatory style training, parents of children
in that group were asked to come to the school to complete follow-up
parent forms.

The Explanatom Style Group

The explanatory style group followed the procedures suggested
in Seligman's book Learned Optimism (1991). During 5 forty-ﬁve
minute sessions, children engaged in exercises to improve their
explanatory style. At each session they learned at least one exercise
designed to improve explanatory style. Between sessions, the
children were asked to complete "homework assignments."

The first session focused on teaching the group the "ABC's"
(Adversity-Belief-Consequence), the relationship between an adverse
event, beliefs about the cause of the event, and emotional and
behavioral consequences. In the next session, the ABC concept was
reviewed, and children were taught to make their own "ABC Records"

in which they articulated personal examples of the connection

35

between events, beliefs, and emotions. Between the second and third
sessions, children were asked to continue with their ABC Records,
and write down ﬁve additional ABC sequences, to be used in the next
session. In the third session, children were taught to dispute their
own beliefs and assumptions that lead to negative emotions. They
were encouraged to generate as many alternatives as possible to
explain why a particular event occurred. In addition, they were
encouraged to look for evidence for and against their belief. After
disputing a belief, they were asked if they felt any emotional change.
Between the third and fourth sessions, children were asked to
dispute one of their own beliefs, and write down how they disputed
it. In the fourth session, children were taught to keep an ABCDE
(Adversity-Belief-Consequence—Disputation-Energization) Record.
Children in the group were asked to bring up a number of adverse
events, and the group worked together to outline the beliefs and
consequences that might follow. The group then offered alternative
interpretations, and speculated as to how alternatives might alter a
person's mood. Between the fourth and ﬁfth sessions, children were
asked to add two complete entries to their ABCDE logs. In the ﬁnal
session, children each chose a piece of paper out of a hat. Each sheet
included an aversive situation (e.g., "The teacher called on you to
answer a question in class, and you blurted out the wrong answer.
The whole class laughed"). Students worked together in groups of
two to go through the entire ABCDE sequence. One child offered
criticisms (negative beliefs) based on the scenario, while the other

child disputed the criticisms.

36

The Relaxation Training Group
The relaxation training group was based on the program used

by James Kahn and his colleagues (1990), and consisted of ﬁve forty-
ﬁve minute sessions. The ﬁrst session focused on the links between
a stressful situation, bodily stress and tension, and depression.
Participants were asked to recall emotional consequences of
situations in which they felt physically tense. In the second session,
children were taught a number of physical relaxation techniques.
For homework, they were asked to practice those techniques before
the next session. During the third session, participants were taught
additional techniques to facilitate physical relaxation. This session
focused on various muscle groups and breathing exercises. Between
the third and fourth sessions, children were asked to write down the
exercise they would be most likely to use in tense circumstances, and
note the emotional consequences that might follow. In the fourth
session, group members learned to use mental strategies for reducing
stress. Between the fourth and fifth sessions, they were asked to
write down at least two adverse situations and how they used (or
could have used) relaxation techniques in dealing with those
situations. The ﬁfth session was a review of the previous four
sessions, and involved practice of mental and physical techniques, as
well as breathing exercises.
The No-Contact Control Group

The children in the no-contact control group did not attend any
of the group meetings over the ﬁve week period. They were asked
to complete post-test and follow-up measures at the same time as

participants of the other two groups. After the follow-up period,

37

children in the no-contact group received explanatory style training.

At the conclusion of this training, they again completed post-test
measures. These children and their parents completed follow-up
measures approximately one month after the ﬁnal group session.
Post-test

At the end of the ﬁnal session, children in the experimental
and placebo groups were asked to write down what they learned in
the ﬁve group sessions. Children from all three groups were asked to
ﬁll out a CDI, a CASQ, the Perceived Competence Scale for Children,
and the Self-Efﬁcacy Scale. Next, participants from all groups were
given alternative versions of the anagram and cross-out tasks.
Follow-Up

One month after the conclusion of the training program,
children attended a follow-up session in which they completed the
CDI, CASQ the Perceived Competence Scale for Children, the Self-
Efﬁcacy Scale, the anagram task, and the cross-out task. They were
also asked whether they spoke to participants in other groups about
the program, and about how much time they spent in discussion
about the group with members of other groups.

At about the same time, parents were invited to attend a
meeting in which they could ask questions about the project, and
give feedback regarding the program's impact, if any, upon their
children. At this time, parents who completed the initial measures
were asked to complete a second CBCL and parent information

form.17 Parents were invited to ask any questions about the project

 

1 7 As before, parents who did not attend the meeting were able to complete the
parent forms over the telephone or by appointment.

38

at this time. All participants were thanked for their participation.
After children in the no-contact group received explanatory style
training, children in the relaxation training group were also invited
to participate in an explanatory style group.
Overview of Parent and Child Measures

In summary, all student participants completed all measures
before training group activities began (pre-test), after the completion
of explanatory style and relaxation training groups (post-test), and
approximately ﬁve weeks after the conclusion of training groups
(follow—up). After follow-up testing was completed, children in the
no-contact group received explanatory style training. Members of
the no-contact group were tested immediately after receiving
explanatory style training (post-treatment) and ﬁve weeks after the
conclusion of explanatory style training (treatment follow-up). Thus,
data from post-treatment and treatment follow-up includes
members of the no-contact group only. Table 2 outlines the schedule

of testing for children in the program.

 

Insert Table 2 About Here

 

Parents of all participants were asked to report on their
children's behavior before the groups began (Time 1) and ﬁve weeks
after the conclusion of groups (Time 2). Parents of no-contact group
members were asked to fill out the measures a third time, ﬁve
weeks after their children ﬁnished explanatory style training (Time

3). Because parents were not on exactly the same time schedule as

39

Table 2
Data Collection Procedures for All Groups

 

 

 

Pre- Training Post- Follow— Training Post- Tx

Group test Type test up Type Tx F.U.
Explanatory

Style X Exp X X --- --— ---
Relaxation

Training X Rel X X - - - - - - - - —
No-contact X - - - X X Exp X X
Note: "X" indicates that data was collected for that group. " Exp"

indicates explanatory style training. "Rel" indicates relaxation

training.

40

their children for completing test instruments, the terms "Time 1,"
"Time 2" and "Time 3" are used, rather than the terms "pre-test,"
"follow-up," and "treatment follow-up." Table 3 shows the data

collection schedule for parent measures.

 

Insert Table 3 About Here

 

41

Table 3
Data Collection Schedule for Parent Measures

 

 

Group Time 1 Time 2 Time 3
(Pre-test) (Follow—up) (Tx Follow-up)
Explan. Style X X ---
Relaxation X X ---
No~contact X X X

 

Note: "X" indicates that data was collected for parents of that

treatment group.

42

Results

Independent samples t-tests were used to compare children
who participated in the program with children who did not
participate or dropped out. No signiﬁcant differences between
participants and non-participants were found on any pre-test
measure.

At the conclusion of their group activities, thirty-three children
responded to a question about what they had learned. In the
explanatory style group, twelve of the ﬁfteen respondents (80%)
gave answers that indicated they understood the basic premises of
their group activities. In the relaxation group, eleven of the thirteen
children (85%) gave adequate responses. All of the ﬁve respondents
from the no-contact group gave satisfactory answers. These
numbers indicate that the majority of children had at least a basic
understanding of the purpose and underlying theme of group
activities.

Persistent Differences Between Depressed and Non-Depressed
Students

Independent samples t -tests revealed that depressed children
responded differently than non-depressed children in several areas
measured at pre-test.

For instance, in all ﬁve domains measured by the Harter
Perceived Competence Scale, depressed children indicated lower
estimations of their own competence. In other words, they were less

pleased by their own behavioral conduct, scholastic abilities, athletic

43

skills, social competence, and attractiveness compared to non-
depressed children. Table 4 shows the mean scores for depressed
and non-depressed students in the overall sample for each area of
perceived competence. Independent samples t - tests were
conducted to determine whether depressed and non-depressed
children differed on each measure at each testing time. Results of

these analyses may also be found in Table 4.

 

Insert Table 4 About Here

 

In addition, at pre—test, depressed children obtained lower
scores than non-depressed children for general self-efﬁcacy and
social self-efﬁcacy. Finally, depressed children exhibited more
pessimistic explanatory styles than non-depressed children. Many of
these differences between depressed and non-depressed children
persisted through post-test and follow-up. Mean self-efﬁcacy scores
and CPCN (composite explanatory style) scores for depressed and
non-depressed children may be found in Table 5. For each of these
measures, t values are also presented to indicate signiﬁcant
differences between depressed and non-depressed children at each

testing period.

 

Insert Table 5 About Here

 

Hyppthesis 1
The ﬁrst hypothesis stated that depressed children in the

explanatory style group would show decreased CDI scores and

Table 4

44

Sample Sizes, Means, Standard Deviations (p, u S_D) and t-tests for differences

between means for Depressed and Non-Depressed Participants

 

 

 

 

 

 

Dependent Variable Time Depressed Non-Depressed t
Behavioral Conduct Pre 23, 31.84, 4.43 28, 41.34, 5.68 6.70a
Post 22, 32.55, 5.26 28, 42.03, 6.27 5.81“1
Follow 21, 34.08, 4.29 27, 40.38, 7.06 3.82"1
Perceived Schol. Competence Pre 23, 14.26, 3.31 28, 20.16, 2.97 6.71a
Post 22, 15.09, 2.86 28, 19.31, 3.88 4.263
Follow 21, 14.95, 3.25 27, 19.52, 4.44 4.11a
Perceived Phys. Competence Pre 23, 13.59, 4.60 28, 19.02, 3.53 4.773
Post 22, 14.81, 2.74 28, 18.36, 3.97 3.73a
Follow 21, 14.36, 3.72 27, 19.16, 3.72 4.80a
Perceived Social Competence Pre 23, 13.86, 4.23 28, 19.06, 3.86 4.594
Post 22, 15.91, 3.39 28, 18.42, 4.12 2.310
Follow 21, 16.89, 3.49 27, 19.58, 4.48 2.270
Perceived Attractiveness Pre 23, 14.77, 4.73 27, 20.70, 3.82 4.90a
Post 22, 16.32, 4.43 28, 21.36, 2.81 4.653
Follow 21, 16.20, 3.62 27, 20.26, 4.12 3.57a

 

ap.300], bp301, cpSDS

45

Table 5
Sample Sizes, Means, Standard Deviations (p, M, S_D) and t-tests for differences
between means for Depressed and Non-Depressed Participants

 

Dependent Variable Time Depressed Non-Depressed 1
General Self-Efﬁcacy Pre 23, 56.65, 7.20 27, 65.54, 8.35 3.998
Post 22, 54.71, 8.14 28, 64.77, 10.10 3.80%1

Follow 21, 56.10, 10.31 27, 59.38, 12.24 0.99

 

Social Self-Efﬁcacy Pre 23, 18.13, 4.34 27, 21.70, 5.33 2.57b
Post 22, 19.82, 4.25 28, 21.59, 3.35 1.65

Follow 21, 20.14, 3.29 27, 21.30, 4.23 1.03

 

Composite Explan. Style Pre 23, 2.45, 3.79 28, 7.65, 5.03 4.20a
Post 22, 1.93, 4.84 28, 5.98, 4.89 2.92b

Follow 21, 2.97, 4.59 27, 5.59, 5.44 1.77

ap_S.OOl,1’pS.Ol,9113.05

46

decreased problem scores as reported by their parents. This
hypothesis was supported by the data. Results of repeated measures
ANOVAs indicated that scores of depressed children in the
explanatory style group declined between pre-test and post-test

[F (1,8) = 8.36, p < .05]. From post—test to follow-up, there was no
signiﬁcant change in depression scores [F (1,8) = .01, p = .92]. Table 6
provides a summary of within-subjects effects for the explanatory
style group between pre-test and post-test. Table 7 includes mean
CDI scores for depressed and non-depressed students in the

explanatory style group.

 

Insert Tables 6 and 7 About Here

 

Parents of depressed children in the explanatory style group
reported a decrease in problem behaviors among their children
between Time 1 and Time 2, although this difference did not reach
signiﬁcance [F(1,7) = 5.31, p = .05 5]. Problem scores for depressed
children in the explanatory style group are provided in Table 8. For
both analyses that were used to test Hypothesis 1, only the data from
depressed participants were included, due to the speciﬁc nature of
the hypothesis. Of course, the exclusion of the remaining data

resulted in a reduction in statistical power.

 

Insert Table 8 About Here

 

47

Table 6

Analysis of Variance for Change in CDI Scores among Depressed
Children in the _E_xplanatory Sgle Group

 

Source g po M_S

PT!
It!

 

Within + Residual 143.78 8 17.97
Pre-test to Post-test 150.22 1 150.22 8.36 .020

48

Table 7

CD1 Scores for Depressed and Non-Depressed Children in the
E_xplanatog Splle Group

 

Time

Pre-test Post-test Follow-up

 

Depressed
p 9 9 9
M 19.89 14.11 13.78
SQ 3.69 6.45 1 1.40
Non-depressed
p 9 9 9
_M_ 3.10 6.1 1 6.03
E 1.55 7.34 6.96

 

49

Table 8

Problem Scores for Depressed Children in the E_xplanatory Sgle
Group

 

 

 

Time
Time 1 Time 2
p 9 8
M .33 .24
g; .26 .27

 

Note. The highest possible problem score is 2, indicating that the
child engaged in all of the negative behaviors "all the time." The
lowest possible score is 0, indicating that the child "almost never"

engaged in any of the problem behaviors listed.

-50

Hyp_othesis 2
The second hypothesis stated that depressed children in the

relaxation group would also experience reductions in depressive
symptoms and problem behaviors. This hypothesis was supported
by the data. A repeated measures ANOVA indicated a signiﬁcant
decrease in depression scores between pre-test and post-test [F (1,4)
= 8.00, p < .05]. From post-test to follow-up, depression scores did
not change signiﬁcantly [P ( 1,3) = .18, p = .70]. Again, the small
sample size for these analyses led to reduced statistical power.
Results of the pre-test to post-test ANOVA are presented in Table 9.
Mean CDI scores for the relaxation training group are given in Table
10.

 

Insert Tables 9 and 10 About Here

 

For depressed children in the relaxation group, no signiﬁcant
change occurred in the amount or severity of behavioral problems as
reported by parents between Time 1 and Time 2 [F (1,4) = 1.78, p=
.25]. Table 1 1 provides mean problem scores for depressed children

in the relaxation training group.

 

Insert Table 1 1 About Here

 

Additional Findings Regarding the First TWO Hymtheses
It should be noted that depressed children in the no-contact

group also experienced a reduction in depressive symptoms between

51

Table 9

Analysis of Variance for Change in CD1 Scores Between Pre-test and
Post-test Among Depressed Children in the Relaxation Training Group

 

Source SS d_f MS

Ir-n
1:3

 

Within + Residual 67.04 4 16.76
Pre-test to Post-test 134.07 1 134.07 8.00 .047

52

Table 10

CD1 Scores for Depressed and Non-Depressed Children in the
Relaxation Training Group

 

Time

Pre-test Post-test Follow-up

 

Depressed
__n_ 5 5 4
M 18.92 11.60 1 1.75
S; 9.05 2.86 3.95
Non-depressed
p 8 8 7
M 2.72 5.13 7.50
S; 2.30 6.69 7.25

 

53

Table 1 1

Problem Behavior Among Depressed Children in the Relaxation

Training Group

 

Time
Time 1 Time 2

 

£1. 5 5
M .23 .15
SD .19 .18

 

54

pre-test and post-test [F (1,7) = 14.82, p< .01]. Table 12 includes
mean CDI scores for depressed and non-depressed children in the

no-contact group.

 

Insert Table 12 About Here

 

Repeated measures ANOVAs conducted with data from the
entire sample indicated that depression scores for the sample as a
whole (including depressed and non-depressed children) decreased
signiﬁcantly from pre-test to post-test [F (1,44) = 6.93, p 5.01] and
from pre-test to follow-up [F(1,42) = 4.50, p < .05]. There was no
signiﬁcant interaction effect of group status over time [pre-test to
post-test: [F (2,44) = .22, p = .81]; pre-test to follow-up: [F (2,42) =
.59, p= .5 6].

No signiﬁcant changes in problem scores were found for these
students between Time 1 and Time 2 [F (1,6) = .01, p= .93]. Mean
problem scores for depressed children in the no—contact group are
located in Table 13.

 

Insert Table 13 About Here

 

Although individual analyses for each treatment group failed to
show signiﬁcant declines in problem scores, a repeated measures
ANOVA using the entire sample uncovered a signiﬁcant decline in
these scores from Time 1 to Time 2 for the overall sample F (1,38) =
11.21, p< .01).

55

Table 1 2

CD1 Scores for Depressed and Non-Depressed Students in the No-
Contact Group

 

Time

Pre-test Post-test Follow-up Post-TX Tx follow-up

 

Depressed
p 9 8 8 4 3
M 20.00 12.32 10.75 8.82 8.00
9 7.07 3.70 6.71 5.03 3.00
Non-depressed
r_1 1 1 1 1 10 5 5
M 2.55 3.47 4.26 8.00 5.60
E 1.29 2.84 4.62 1 1.77 7.60

 

56

Table 1 3
Problem Scores for Depressed Children in the No—Contact Group

 

 

 

Time
Time 1 Time 2 Time 3
8 8 3
.38 .36 .29

'8 I: It:

.30 .23 .28

 

57

Pearson correlation coefﬁcients were examined to determine
the relationship between explanatory style and depression. Analyses
revealed that explanatory style at pre-test was signiﬁcantly
correlated with CD1 scores at pre—test (r = -.54, p < .001), post-test (r
= -.55, p < .001), and follow-up (r = -.54, p < .001). Post-test CPCN
scores were not as strongly associated as pre-test CPCN scores with
CD1 scores at post-test (r = -.3 1, p < .05) or follow-up (r = -.16, p=
.27). In the same way, follow-up CDI scores were not as highly
correlated with CPCN scores at follow-up (r = -.39, p < .01) as they
were with pre-test CPCN scores.

Hyppthesis 3

The third hypothesis predicted that students in the explanatory
style group would show more optimism than students in the other
two groups at post-test and follow-up. Results of independent
samples t -tests did not support this hypothesis. Mean CPCN scores

for each group are presented in Table 14.

 

Insert Table 14 About Here

 

Additional Findings Regarding Hypothesis 3
It should be noted that pre-test CPCN scores suggested that the

 

no-contact group was the most optimistic, and the explanatory style
group was the least optimistic of the three groups. This difference
did not attain signiﬁcance [t (36) = -1.86, p = .072]. At follow-up,
explanatory style group members attained the highest CPCN scores,
whereas the no-contact group members obtained the lowest. This

trend was also non-signiﬁcant [t (34) = 1.23, p= .23].

Table 1 4

58

E_xplanatory Style and Group Membership

 

 

 

 

 

Time h M. S).
Explanatory style group
Pre-test 18 3.75 5.26
Post-test 1 8 4.41 5.15
Follow-up 1 8 5.22 5.27
Relaxation training group
Pre-test 13 5.35 5.88
Post-test 1 3 4.5 1 5.48
Follow-up 1 2 5.00 6.59
No-contact group
Pre-test 20 6.67 4.44
Post-test 1 9 3.78 5.40
Follow-up 1 8 3.29 4.07
Post-treatment 9 5.21 5.62
Treatment Follow-up 3.57 5.97

59

A repeated measures AN OVA was conducted for the entire
sample to assess changes in CPCN scores over time. In the overall
sample, there were no signiﬁcant changes in explanatory style from
pre-test to follow-up [F(1,42) = 1.22, p = .28]. However, there was a
signiﬁcant interaction effect indicating that the three treatment
groups showed different patterns of change in CPCN scores during
this time period [F(1,42) = 4.75, p _<.01]. Whereas the relaxation and
no-contact groups showed decreased optimism between pre-test and
follow-up, the explanatory style group experienced some increase in
optimism. Figure 1 shows the change in explanatory style for the

three treatment groups over time.

 

Insert Figure 1 About Here

 

Separate analyses for each treatment group revealed that the
no-contact group showed a signiﬁcant decrease in optimism between
pre-test and post-test [F (1,17) = 6.62, p < .05], and between pre-test
and follow-up [F (1,17) = 15.36, p < .001]. The other two treatment
groups' changes in overall optimism were not found to be signiﬁcant.

Upon closer examination of the components of the CPCN score,
it was found that the no-contact group was signiﬁcantly more
optimistic in the face of negative events (CN) at pre-test compared to
the explanatory style group [t (36) = 3.54, p < .001] and the
relaxation training group [t (31) = 2.70, p _<.01]. These differences
disappeared at post-test and follow-up. There were no signiﬁcant
differences between the three groups in explanatory style for

positive events (CP) at pre-test or post-test. However, at follow-up,

6O

 

—D— Explan.Style

Q
" ‘. ........ O ........ Relaxation

- - - - Q- - - - No-Contact

CPCN Scores
0.
m

 

03--
o,

 

 

 

. 'I-1

l’reotest ..

Post-test
Follow-up 2
Post Tx ~

Tx Follow-up -1

Time

Figure 1. Composite explanatory style (CPCN) by treatment group at
each data collection point.

61

the explanatory style group obtained higher CP scores than the no—
contact group [t (34) = 2.47, p < .05]. A repeated measures ANOVA
indicated that the change in CP scores among the explanatory style
group between pre-test and follow-up was signiﬁcant [F (1,17) =
6.28, p < .05]. In addition, the no-contact group showed a signiﬁcant
decline in optimism for negative events between pre-test and post-
test [F(1,17) = 8.85, p < .01] and between pre-test and follow-up [F
(1,17) = 33.63, p < .001]. Figures 2 and 3 show the changes in CP and

CN scores over time for the three treatment groups.

 

Insert Figures 2 and 3 About Here

 

Hypothesis 4

According to the fourth hypothesis, members of the
explanatory style group were expected to show greater perceived
competence and a stronger sense of self-efﬁcacy than the other two
treatment groups at post-test and follow-up. This hypothesis was
not supported. Independent samples t-tests did not indicate
signiﬁcant differences between the explanatory style group and the
other two treatment groups at any time on any measure of perceived
competence (Perceived Competence Scale for Children; Child Behavior
Checklist, Social Competence Subscale) or self-efﬁcacy.

Pearson correlation coefﬁcients indicated that explanatory style
at pre-test was correlated with a measure of general self-efﬁcacy at
pre-test (r= .54, p < .001), post-test (r= .43, p < .01), and follow-up (r
= .41, p < .01 ). Pre-test CPCN scores were also related to social self-
efﬁcacy at post-test only (r = .3 1, p < .05). CPCN scores at post-test

 

 

 

62

 

 

 

 

l 0
1 0—0/0
h .71.Q‘
9 '1 ....... 8,3"
m ...... 0., --
2 l. 0' .' - 0
g c: “\\ “\ '.o.
2 8 _4 0 ’0'" —-C}— Explan.Style
.g '
3 ........ o cccccccc RciaA-diiuu
z . 1
0 0.
a 7 .J .' -- ------0 bio-Contact
8.
E .'
c
U ,‘
6 -‘ .'
O
5 I I I I 1
5'5 9. E g .5.
o = -'
a. co. {2,
Time is

 

Note.- Higher scores indicate increased pessimism in the face of
negative events.

Figure 2. Explanatory style for negative events (CN) by treatment

group at each data collection point.

63

 

 

 

 

 

15
14.5 -
«"3
5 14 4
é’ —CI-— Explan.Style
0
>
:5 13 5 .1 ........ o ........ Relaxation
o
a.
,3 ,Q, --- - O---- No-Contact
§ 13 - - '
5 '
C.) ,' ‘.‘
12.5 a ______ O """" <3 '.
O"
b
12 T F I I t
a 28: s- : 3
Time :5.

Figure 3. Explanatory style for positive events (CP) by treatment

group at each data collection point.

64

were signiﬁcantly correlated with general self-efﬁcacy scores at
post-test (r = .37, p < .01) and follow-up (r= .39, p < .01), and with
social self-efﬁcacy at post-test only (r= .41, p < .01). Follow-up CPCN
scores were signiﬁcantly associated with both general self-efﬁcacy at
follow-up (r = .52, p < .001) and social self-efﬁcacy at follow-up (r=
.72, p<.001).

In regard to the relationship between perceived competence
and explanatory style, it was found that CPCN scores at pre-test were
signiﬁcantly correlated with all indices of perceived competence at
pre-test, post-test, and follow-up. These correlations ranged from
.29 to .62. With the exception of the follow-up behavioral conduct
score, pre-test CPCN measures were more highly correlated than
CPCN scores at post—test and follow-up with all indices of perceived
competence at pre-test, post-test, and follow-up. A correlation
matrix including correlations between CPCN scores and perceived

competence scores is provided in Table 15.

 

Insert Table 15 About Here

 

Hymthesis 5
Students in the explanatory style group were expected to show

more persistence behaviors than members of the other two
treatment groups after receiving explanatory style training. The
data did not support this hypothesis. No signiﬁcant differences were
found among the three groups in terms of numbers of correct

responses or amount of time spent in the two tests of persistence.

Table 15

65

Correlations mtween CPCN Scores and Indices of Perceived Competence

 

CPCN (Pre)

CPCN (Post)

CPCN (Follow-up)

 

Behavioral Conduct

 

Pre .6 1*** .35** .3 2*
Post .50*** .37** .47***
Follow-up .3 9** .50*** .49***
Scholastic
Pre .60*** .25 .33*
Post .5 2*** .41** ' .48***
Follow-up .49*** .47*** .43**
Social
Pre .3 3* .26 -.01
Post .42** .21 .21
Follow-up .3 9** .12 .26
Physical
Pre .47*** .05 .05
Post .5 9*** .09 .18
Follow-up . 6 2 * * * .20 .2 8*
Attractiveness
Pre .5 3*** .18 .15
Post .29* .17 .15
Follow-up .3 4* .06 .18
*p < .05. **p < .01. ***p_<.001.

Table 15

65

Correlations betwgp CPCN Scores and Indiges of Perceived Competence

 

CPCN (Pre)

CPCN (Post)

CPCN (Follow-up)

 

Behavioral Conduct

 

Pre .6 1*** .3 5** .3 2*
Post .50*** .37** .47***
Follow-up .3 9** .50*** .49***
Scholastic
Pre .60*** .25 .3 3*
Post .5 2*** .41** ' .43***
Follow-up .49*** .47*** .43**
Social
Pre .33* .26 a01
Post .42** .21 .21
Follow-up .3 9** .12 .26
Physical
Pre .47*** .05 .05
Post .5 9*** .09 .18
Follow-up .6 2*** .20 .2 8*
Attractiveness
Pre .5 3*** .18 .15
Post .29* .17 .15
Follow-up .3 4* .06 .18
*p < .05. **p < .01. ***p<.001.

66

Additional Findings Regarding Hypothesis 5
CPCN scores were not signiﬁcantly related to scores for either

of the tasks of persistence, with the exception of a signiﬁcant
association between CPCN scores at follow-up and cross-out scores at
follow-up (r= .34, p<.05).

The two tests of persistence, an anagram task and a cross-out
task, yielded very different results for depressed and non-depressed
children. On the cross-out task, depressed children obtained higher
scores than non-depressed children. This difference was not
signiﬁcant at pre-test [t (49)= -1.30, p = .20], but did reach
signiﬁcance at post-test [t (48) = -2.54, p < .05] and follow-up [t (46)
= -2.43, p < .05]. Means for depressed and non-depressed children in

the overall sample may be found in Table 16.

 

Insert Table 16 About Here

 

Levene's test for equality of variances showed signiﬁcantly
different variances for depressed and non-depressed children on the
cross-out measure. Variances for the non-depressed group were
signiﬁcantly larger than variances for the depressed group at post—
test [F (1,48) = 7.98, p < .01], and follow-up [F (1,46) = 12.77, p _<
.001]. This discrepancy indicated that non-depressed students
showed a signiﬁcantly wider range of scores compared with the
depressed group.

In contrast to their performance on the cross-out task,

depressed children gave signiﬁcantly fewer correct responses than

67

Table 1 6
Cross—01g Persistence of Depressed and Non-Depressed Participants

 

 

 

 

 

 

Time I_1 M .32 Ld
Depressed children
Pre-test 23 80.52 7.36
Post-test 22 82.36 6.82
Follow-up 2 1 83.00 7.34
Non-depressed children
Pre-test 28 76.1 1 14.77
Post-test 28 7 1.79 20.67
Follow-up 27 73.04 19.57
Total sample
Pre-test 51 78.10 12.11 -1.30
Post-test 50 76.44 16.84 -2.54*
Follow-up 48 77.40 16.12 -2.43*
*p < .05.

 

C1 t values compare depressed and non-depressed students.

68

non-depressed children on the anagram task at pre-test [t (49) =
2.15, p< .05], and follow-up [t (46) = 3.31, p < .01]. (At post-test, this
discrepancy did not attain signiﬁcance [t (48) = 1.33, p = .19]. Mean
scores for depressed and non-depressed children on the anagram
task may be found in Table 1 7.

 

Insert Table 17 About Here

 

Hyp_othesis 6
After receiving explanatory style training, the no-contact group

was expected to show fewer depressive symptoms, greater perceived
competence, higher self-efﬁcacy, and more optimistic explanatory
style than the relaxation training group. This hypothesis was
somewhat difﬁcult to test because of the high attrition rate among
members of the no-contact group between follow-up and post-
treatrnent, and the relatively small numbers of relaxation group
members. The current data do not offer corroboration for this
hypothesis.

Independent samples t-tests were conducted comparing the
relaxation group at post-test and follow-up with the no-contact
group at post-treatment and treatment follow-up. No signiﬁcant
differences were found between the two groups in regard to
depression scores, perceived competence, self-efﬁcacy, or

explanatory style.

69

Table 1 7
Anagpa_m Persistence of Depressed and Non-Depressed Participants

 

 

 

 

Time _Ii M E I. d
Depressed children
Pre-test 23 3.61 2.61
Post-test 22 3.00 2. 1 2
Follow-up 2 1 3.33 2.58
Non-depressed children
Pre-test 28 5.00 1 .85
Post-test 2 8 3.79 2.04
Follow-up - 26 5.70 2.37
Total sample
Pre-test 5 1 4.37 2.31 2.15*
Post-test 50 3.44 2.09 1 .33
Follow-up 47 4.67 2.7 1 3.3 1 **

 

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

 

d t values compare depressed and non-depressed students.

70

Discussion

Contrary to predictions, group treatments did not lead to
signiﬁcant differences among the three experimental groups on a
number of measures of emotional functioning. However, increases in
optimism for positive events among children in the explanatory style
training group lend some support to the notion that interventions can
lead to change in the explanatory style of pre—adolescent children.

Undoubtedly, the most striking ﬁnding of this report is that
depression in children seems to color almost every aspect of their
emotional functioning. At pre-test, depressed children were more
likely than non-depressed children to have: (a) lower perceived
competence in all ﬁve of the domains measured by the Harter
Perceived Competence Scale (behavioral conduct, cognitive abilities,
social competence, athletic competence, and attractiveness), (b) lower
general and social self-efﬁcacy scores, and (c) a more pessimistic
explanatory style. These ﬁndings mirror those of Kaslow, Rehm, and
Siegel (1984), in which depressed children showed more pessimistic
explanatory style, lower self-esteem, and lower expectations for
success on a cognitive task compared to non-depressed children.

Obviously, deﬁcits of these kinds may result in signiﬁcant
difﬁculties with peer relations as well as reduced productivity in the
classroom. In turn, unsatisfying peer relations and academic
difﬁculties may lead to further increases in depression level. In this
way, childhood depression may result in a "snowball effect" in which
children fall behind their classmates in terms of social and academic

competencies, and thus become increasingly depressed.

71

In particular, the current investigation indicates that depressed
children do not have as much conﬁdence in their abilities as non-
depressed children. Moreover, these children may not have gained
the necessary skills to deal with difﬁcult situations, due to their lack
of experience in relying on their own abilities to make friends,
behave appropriately, or solve problems. This lack of conﬁdence and
skills may translate into a learned helplessness mindset, in which
trying to improve a situation is seen as an exercise in futility.

Changes in Depression Status and Behavior Problems
The ﬁrst hypothesis stated that depressed children in the

 

explanatory style group would show less depression and fewer
behavioral problems after the conclusion of their ﬁve-week training
course. Although depression among this group decreased over time,
signiﬁcant reductions in problem behaviors did not emerge.

There have been numerous investigations in which
maladaptive explanatory style has been associated with depressive
symptomatology in both children and adults (Garber et al., 1993;
Kaslow et al., 1994; McCauley et al., 1988; Nolen-Hoeksema et al.,
1986; Nolen-Hoeksema et al., 1992; Rizley, 1978; Seligman et al.,
1988) The current investigation also indicated an inverse
relationship between optimism and depression, although the
direction of causality between the two variables remains unclear.

A decline in reported behavioral problems was also predicted
for this group. Decreases in depression scores were expected to
result in fewer externalizing and "acting out" behaviors, as viewed by
the parents. However, signiﬁcant decreases in CBCL problem scores

did not emerge at Time 2.

72

Earlier research indicated that relaxation training groups could
also have a positive impact on depression levels (Kahn et al., 1990).
This hypothesis was supported in the current study, as depressed
children in the relaxation group experienced decreases in depression
scores. However, depressed children in this group did not show
signiﬁcant declines in CBCL problem scores.

One reason for a lack of signiﬁcant decreases in parent reports
of behavioral problems for each treatment group may be the short
duration of group activities. It may take time for children to reduce
problem behaviors following a reduction in depressive
symptomatology. Alternatively, it may take time for parents to
notice behavioral changes in their children, and to accept them as
"changes" rather than temporary ﬂuctuations in the child's usual
behavior.

Perhaps more importantly, the small numbers of children
included in analyses for each separate group seemed to mask
reductions in problem behaviors. Even though all groups showed
some non-signiﬁcant reductions in CBCL problem scores between
Time 1 and Time 2, no group's decline reached signiﬁcance. This lack
of statistical power to ﬁnd changes in CBCL problem scores among
each separate treatment group was underscored by the ﬁnding of a
significant decline in problem scores in the overall sample. It is
recommended that future investigations include larger numbers of
children in each treatment group in order to increase statistical
power. In addition, an increase in the potency of the interventions
(which might be achieved by increasing the number of group

sessions) would also have the effect of boosting power.

 

 

 

73

It should be noted that the no-contact group experienced a
signiﬁcant decrease in depression scores, but no signiﬁcant change in
reported behavioral problems. Thus, data from the three groups do
not indicate that training is directly responsible for a decrease in
depression scores over time. However, it is possible that children in
all three groups derived some beneﬁt from being part of the STARS
program, in that each student was known (by name) to the
experimenter, and was recognized within the school as a program
participant. Even children in the no-contact group were able to look
forward to participation in the groups and an awards banquet at the
conclusion of the program. This situation may have contributed to a
feeling of anticipation and a reduction in depression scores and
problem scores among the sample as a whole.

Another possibility is that the reduction in depression scores
was simply a case of regression to the mean, suggesting that
depression status in children of this age group is somewhat unstable.
However, an examination of the data shows a signiﬁcant discrepancy
between depressed and non-depressed children in CD1 scores at pre-
test, post-test, and follow-up. This persistent difference between
depressed and non-depressed students indicates that depression is a
somewhat chronic problem among many of the children in the
depressed classiﬁcation. Moreover, persistent differences between
depressed and non-depressed children in other areas of functioning
suggest some real differences between depressed and non-depressed
children.

The decrease in depression scores among children in the no-

contact group may also be related to their relatively high composite

74

CASQ scores (CPCN scores) at pre-test. Although the direction of
causality linking depression and explanatory style is unclear at
present (Garber et al., 1993; Lewinsohn et al., 1981; Nolen-Hoeksema
et al., 1986), initial optimism among the no-contact group may have
facilitated a reduction of depressive symptomatology between pre-
test and post-test.

In regard to parent reports of behavioral problems, it may be
useful to conduct future research using larger samples of children in
each treatment group in order to increase statistical power in
analyses of each separate treatment condition. Moreover, longer
interventions may provide additional time for children to change
their behavior, and for parents to perceive these changes as stable.
Explanatog; Style

Children in the explanatory style training group did not have
signiﬁcantly more optimistic explanatory styles than children in the
other two groups at post-test or follow-up. However, the
explanatory style group did evidence signiﬁcant improvement in
composite positive explanatory style (CP) scores over time. In
contrast, the no-contact group evidenced a signiﬁcant increase in
pessimism for negative events and a signiﬁcant decrease in overall
optimism over time. Thus, an interaction effect emerged such that
different treatment groups experienced different patterns of change
in CPCN scores over time (Figure 1 shows the differential patterns of
change in CPCN scores among the three groups.)

The School Environment
It is not clear why the no-contact group showed decreased

optimism over time. It is possible that the special challenges faced

75

by students in this community made it difﬁcult to maintain an
optimistic style, particularly in the face of poverty and pressing
community-wide health concerns. Over the course of the program,
there were a number of indications that many families in the
community were struggling ﬁnancially. For example, forty-ﬁve
percent of S.T.A.R.S. parents reported an annual income of less than
$ 10,000. Furthermore, many children in the program took
advantage of a school breakfast program.

In addition, there seemed to be a high incidence of physical
illness in the community. Throughout the course of the program, a
number of children "dropped out" due to illnesses lasting for more
than two weeks. Several children also reported serious illness among
their parents.

The 1995-1996 school year may have been especially stressful
in the classroom, as this middle school received the lowest basic
skills test scores of all schools in the state, for the second year in a
row. Obviously, this situation put additional pressure on teachers to
improve student performance.

Still, students had a number of resources within the school. For
instance, many students participated in the Helping One Student to
Succeed (HOSTS) program, a mentoring program in which children
were paired with adults in the community to receive one-on-one
help with language arts, including reading and writing skills.
Students in the HOSTS program were expected to attend the program
one hour daily, during school hours. In addition, the school had
excellent computing facilities, in which students had the opportunity

76

to learn the basics of word processing and other computer functions.
In general, school facilities were clean and in good condition.
Promoting Exgplanatog Sgle

The ﬁndings of the present study suggest that it is worthwhile
to begin promoting optimism in pre-adolescents, since they seem
capable of changing explanatory style during this developmental
period. In fact, the majority of students in the explanatory style
group were able to articulate the purpose of the group, indicating
that they had at least a basic understanding that there are many
ways of looking at situations, and that beliefs affect the way we feel.
Emphasis on explanatory style in this age group may increase the
chances of developing more enduring optimism.

However, predicted changes in areas such as persistence,
perceived competence and self-efﬁcacy did not appear, even with
increases in optimism for positive events. It makes sense that
students who begin to believe in their own ability to make good
things happen (i.e., those who make internal attributions for positive
events) would be expected to show conﬁdence in many of their skills
and abilities. Yet, it may take time for these individuals to change
behavior patterns and begin to have conﬁdence in their own skills.
Further, the many elements of a child's living situation (e.g., the
classroom, peer relations, and home life) might not reinforce, or even
support, the child's new optimistic thinking. Finally, time-limited
group activities may not provide the child with sufﬁcient
opportunity to internalize the connection between effort and
achievement. If this connection is not made, increased persistence is

not likely to be obtained.

77

Optimism for Positive and Negative Events
Speciﬁcally, it appears that the explanatory style training

approach was successful in increasing optimism for positive events.
However, optimism in the face of negative events did not change
signiﬁcantly after training. Perhaps explanatory style training acted
as a protective factor to mitigate the effects of negative events on
explanatory style. In this case, the no-contact group may have
become increasingly pessimistic in the face of negative events,
whereas the explanatory style group did not show signiﬁcant change
in that area. Future interventions may beneﬁt from an additional
focus on problem-solving skills for dealing with difﬁcult situations in
positive ways.

It should be noted that many of the stressors faced by these
students cannot be "wished away" by optimism and positive
thinking. Obviously, these children cannot control their families'
economic situations or the behavior of their parents or any number
of other events. Thus, future researchers may wish to include a
"coping" component to interventions in order to help students deal
with some of these uncontrollable stressors.

One model for working with children in this age group was
provided by the Penn-Prevention Program (Jaycox et al., 1994).

The Penn program included weekly 90-minute meetings in which
children participated in cognitive training (including explanatory
style training), social problem-solving training, or a combined
approach including both cognitive and problem-solving aspects.
Children in treatment groups experienced a reduction of depressive

symptoms compared to control children. However, parents did not

78

report a corresponding decrease in behavioral problems following
group activities. Furthermore, there was no signiﬁcant change
among these children in CPCN scores at the conclusion of the training.
Results of the current investigation suggest that a greater focus on
explanatory style may result in increased optimism, and perhaps a
reduction in behavioral symptoms as well.

Perceived Competence and Self-efﬁcacy

The fourth hypothesis stated that children in the explanatory
style group would obtain signiﬁcantly higher scores than children in
the other two groups in perceived competence and self-efﬁcacy. Yet,
as stated earlier, no signiﬁcant differences were found among the
three treatment groups on these measures. Perhaps interventions
that focus solely on explanatory style are not sufﬁcient to alter
functioning in these areas. Alternatively, this type of intervention
may have been adequate, but of insufﬁcient duration to signiﬁcantly
modify perceptions of competence and self-efﬁcacy. Additional
sessions might have increased students' conﬁdence in themselves
and helped them to see themselves as capable of inﬂuencing future
events by their actions in the present.

Furthermore, time and context may play an important role in
changing these variables. Reinforcement of positive explanatory
style at home and at school may help students increase perceptions
of their own competence. In turn, this type of contextual change
may help children to associate effort with positive outcomes, by
viewing good events as determined by internal, stable, and global

Causes.

 

79

This investigation does provide some evidence that self-
efﬁcacy and explanatory style are related. For example, at follow-up,
children's CPCN scores were relatively highly correlated with their
general self-efﬁcacy (r= .52, p < .001) and social self-efﬁcacy (r= .72,
p < .001) scores.

In addition, some areas of perceived competence, such as
behavioral conduct and scholastic competence, were found to be
signiﬁcantly related to explanatory style. Furthermore, CPCN scores
at pre—test were signiﬁcantly associated with all indices of perceived
competence at all testing sessions. Thus, it is possible that enduring
changes in explanatory style will lead to a heightened sense of self-
efﬁcacy and increased perceived competence.

Persistence

Persistence is based on the hope that one's efforts will be
rewarded at a later point in time. Thus, it requires some delay of
gratiﬁcation. The existing research on delay of gratiﬁcation is helpful
in elucidating some of the beneﬁts of persistence among children.

For example, in a review of the literature, Mischel, Shoda, and
Rodriguez (1989) concluded that the ability of children to delay
gratiﬁcation is predictive of later social and cognitive competence, as
well as enhanced coping skills when dealing with stress and
frustration. It makes sense that children who persist in their
relationships with people (rather than "discarding" friendships when
they seem unrewarding) may develop a larger support network than
children who are less persistent. In the same way, task persistence

may result in higher levels of academic achievement.

 

 

 

80

The explanatory style group was expected to show more
persistence behaviors than the other two groups on both persistence
tasks (Hypothesis 5). However, the number of correct responses and
the amount of time taken to complete these tasks were unrelated to
treatment group membership. Thus, the current investigation does
not support the notion that explanatory style training is an effective
method for increasing persistence.

The perception that "persistence pays off" may be a necessary
prerequisite to signiﬁcant increases in persistence behaviors, and it
may take more than ﬁve weeks for students to arrive at this
realization. Future interventions should increase the number of
sessions, and seek to involve more students in order to obtain more
conclusive results.

Persistence and Depression

Findings on measures of persistence were mixed with regard to
depression status. While depressed children consistently performed
better than non-depressed children on a cross-out task, the reverse
was found for an anagram task. These outcomes may be related to
the different nature of the two tasks. Whereas the cross-out
measure was not difﬁcult, and did not require higher level cognitive
processing, the anagram task was certainly more cognitively
challenging, as the children attempted to arrange a series of letters to
form a familiar word.

Perhaps depressed children saw the cross-out task as an
opportunity for success, one which did not make excessive cognitive
demands. On the other hand, the anagram problems may have

seemed too difficult for depressed children to even seriously

 

 

81

attempt. During the testing itself, it was apparent that many of the
students did not take the anagram task seriously, but instead
returned it quickly, either with few written responses, or with a
number of "nonsense" responses in which the letters were
rearranged but did not form words.

It appears that optimal performance on tasks of persistence
may be the result of a delicate balance of cognitive challenge and
opportunity for success. This balance is likely to differ from child to
child.

Similar to the results of this study, other investigators have
also demonstrated that depressed children have difﬁculty with
complex cognitive tasks such as anagrams and block design (Kaslow
et al., 1983; Ward, Friedlander and Silverman, 1987). In addition,
Laurer, Giordani, Boivin, & Halle (1994) found that depressed
children do more poorly than non-depressed children on higher
order memory tasks, but perform at about the same level as non-
depressed children on automatic memory tasks. In the same way,
Nadine Kaslow and her associates (1984) found that the performance
of depressed children was impaired in some cognitive tasks, such as
block design, coding, and digit span, but was comparable to the
performance of non-depressed children on a test of vocabulary. The
results of these studies and the present investigation suggest that all
tasks of persistence do not yield the same ﬁndings, especially when
depression status is considered. Clearly, the cognitive complexity of
a task is also a key factor in determining how a child will perform.

Results of a study by Diener and Dweck (1978) shed some light

on differences between "helpless" and "mastery-oriented" children (a

82

distinction which is likely to be related to depression status).
According to their study, when confronted with failure, helpless
children engaged in self-talk attributing difﬁculties to a lack of
ability (internal attribution for failure), whereas mastery-oriented
children's self-talk focused upon self-instruction and a search for
solutions. Whereas negative self-talk may result in a decrease in
attention and effort, solution-oriented talk may result in increased
attention, effort, and performance. It is possible that depressed
children in the current investigation engaged in discouraging self-
talk that led them to stop working on the anagrams task
prematurely.

In a related study, Diener and Dweck (1980) found that
helpless children viewed (academic) successes as external and
unstable. Clearly, these beliefs are not conducive to the development
of persistence. It is reasonable to speculate that children can gain
conﬁdence in themselves when they begin to view their successes as:
1) related to their own efforts and abilities (internal), 2)
generalizable to other areas (global), and 3) likely to continue
(stable). Future interventions that involve more children and last
longer than ﬁve weeks may result in increased persistence behaviors
as well as an enduring optimism for both positive and negative
events.

The No—contact Group

Because of the small number of no-contact group members who
actually completed explanatory style training, the sixth hypothesis
was difﬁcult to test. Children in the no—contact group were expected

to show lower depression scores, and higher scores in perceived

 

83

competence, self-efﬁcacy, and explanatory style, as compared with
members of the relaxation training group. Following treatment, the
no-contact group did not differ signiﬁcantly from the relaxation
group in terms of perceived competence, self-efﬁcacy, or explanatory
style.

Future Directions

Context plays a large role in behavior, and can facilitate or
obstruct change. Children who participated in group activities
experienced some change in context, but only for a total of ﬁve hours
over a ﬁve-week period. Still, following explanatory style training,
children were able to increase their reported levels of optimism on
the Children's Attributional Style Questionnaire. Future research that
seeks to involve parents and teachers more fully in changing
explanatory style may result in corresponding changes in other
aspects of emotional functioning such as depression, self-efﬁcacy,
and perceived competence. In addition, this reinforcement of
optimistic thinking may lead to increased persistence on cognitively
demanding tasks.

The current training program focused solely on improving the
explanatory style training of pre-adolescent children. Such a focus
has both costs and beneﬁts associated with it. One advantage was
that basic concepts regarding explanatory style (e.g., the idea that
beliefs are related to feelings) were reviewed numerous times within
the five week course, so that children were not likely to miss the
central ideas of the program. In addition, the data from this program
suggest that explanatory style can be taught to children fairly

84

directly, using exercises and discussions similar to those described by
Seligman (1991).

Still, future programs may beneﬁt from more attention to the
challenges of growing up in a time and place where children are
exposed to poverty and other mental and physical health issues.
Decreases in optimism in the face of negative events may be
combated with effective problem-solving skills, in addition to
explanatory style training. A coping skills training component may
also be useful to children facing various uncontrollable stressors.

Although children who received explanatory style training
showed increased optimism, they did not show changes in other
areas that would distinguish them from the other two treatment
groups. Further research should be done, in which interventions last
longer than ﬁve weeks, and allow parents and/ or teachers to
participate along with the students. These considerations may
increase the potency of the intervention and actually change the
context in which the children are living.

In addition, changes in emotional functioning may take time to
surface. In other words, children may require more than ﬁve weeks
to internalize a sense of conﬁdence in their own abilities. Clearly,
longer interventions may reveal whether an " incubation peri " is
needed for children to show changes in behavior and emotional
functioning following decreases in depression level and/ or increases
in optimism. Finally, as stated earlier, larger samples will reduce
sampling error, and will facilitate stronger conclusions about the

beneﬁts and limitations of this type of intervention.

LIST OF REFERENCES

85

LIST OF REFERENCES

Abramson, LY., Seligman, M.E.P., & Teasdale, J. (1978).
learned helplessness in humans: Critique and reformulation. Journal

‘ of Abnormal Psychology, _81 , 32-48.

Achenbach, T.M., & Edelbrock, CS. (1983). Manual for the Child
Behavior Checklist and Revised Child Behavior Proﬁle. Burlington,
University of Vermont.

Alloy, L.B., Abramson, L.Y., Metalsky, G.I., & Hartlage, S. (1988).
The hopelessness theory of depression: Attributional aspects. British

lournal of Clinical Pychology, 27, 5-21.

Andrews, G.R., & Debus, RL. (1978). Persistence and the causal
perception of failure: Modifying cognitive attributions. Ioumal of

Educational Psychology,JQ(2), 1 54-166.

Beck, A.T., Rush, A.J., Shaw, B.F., & Emery, G. (1979). Cognitive

therapy of depression: A treatment manual. New York: Guilford
Press.

Beck, A.T., & Weishaar, M. (1989). Cognitive therapy. In A.
Freeman, K.M. Simon, LE. Beutler, & H. Arkowitz (Eds.),
Comprehensive handbook of cognitive therapy (Pp. 21-36). New
York: Plenum Press.

Beecher Community School District (1990). 1990 Census
Tracking Data.

Boyle, M.H., Offord, D.R., Hofmann, H.G., Catlin, G.P., Byles, J.A.,
Cadman, D.T., Crawford, J .W., Links, P.S., Rae-Grant, N.I., & Szatmari, P.
(1987). Ontario child health study. Archives of General Psychiatry,
44, 826-83 1.

Braswell, L. (1991). Involving parents in cognitive-behavioral
therapy with children and adolescents. In P. Kendall (Ed.), Child and
adolescent therapy (pp. 316-351). New York: The Guilford Press.

86

Butler, L., Miezitis, S., Friedman, R., & Cole, E. (1980). The effect
of two school-based intervention programs on depressive symptoms

in preadolescents. American Education Research loumal,ﬂ, 1 1 1-
1 1 9.

Carey, M.P., Faulstich, M.E., Gresham, F.M., Ruggiero, I.., &
Enyart, P. (1987). Children's depression inventory: Construct and
discriminant validity across clinical and nonreferred (control)

populations. lournal of Consulting and Clinical Psychology, 5 5(5),
755-761 .

Chapin, M., & Dyck, DC. (1976). Persistence in children's
reading persistence as a function of n length and attribution

retraining. lournal of Abnormal Psychology,§5(5), 5 1 1-5 1 5.

Cole, D.A., & Turner, J .E., Jr. (1993 ). Models of cognitive
mediation and moderation in child depression. lournal of Abnormal

Psychology, 102, 271-281.

Cronbach, L.J. (1951 ). Coefﬁcient alpha and the internal
structure of tests. Psychometrika, 16, 297-309.

Diener, C.I., & Dweck, CS. (1978). An analysis of learned
helplessness: Continuous changes in performance, strategy, and
achievement cognitions following failure. lournal of Personality and

Social Psychology, 36(5), 451-462.

Diener, C.I., & Dweck, CS. (1980). An analysis of learned
helplessness: II. The processing of success. journal of Personalig
and Social Psychology, 39(5), 940-95 2.

Elias, M.J., Gara, M., Ubriaco, M., Rothbaum, P.A., Clabby, J .F., &
Schuyler, T. (1986). Impact of a preventive social problem solving
intervention on children's coping with middle-school stressors.

American lournal of Communig Psychology,_1_4, 259-275.

Ellis, A. (1962). Reason and emotion in psychotherapy.
Seacaucus, NJ: Lyle Stuart.

Emery, R.E., & O'Leary, K.D. (1982). Children's perceptions of
marital discord and behavior problems of boys and girls. Journal of

Abnormal Child Psychology, 10, 1 1-24.

87

Fine, S., Forth, A., Gilbert, M., & Haley, G. (1991). Group therapy
for adolescent depressive disorder: A comparison of social skills and

therapeutic support. Journal of the American Academy of Child and
Adolescent chhiatgﬁﬁ, 79-85.

Fleming, J .E., & Offord, DR. (1990). Epidemiology of childhood
depressive disorders: A critical review. Journal of the American

Academy of Child and Adolescent Psychiatg, 29(4), 571-5 80.

Fowler, J.W., & Peterson, P.L. (1981). Increased reading
persistence and altering attributional style of learned helpless
children. Iournal of Educational Psychology,_13_(2), 25 1-2 60.

Garber, J., Weiss, B., & Shanley, N. (1993). Cognitions,
depressive symptoms, and development in adolescents. Journal of
Abnormal P cholo 102 (1), 47-57.

Harter, S. (1982). The Perceived Competence Scale for Children.
Child Development, 53, 87-97.

Harter, S. (1987). The determinants and mediational role of
global self-worth in children. In N. Eisenberg (Ed.), Contemmm

topics in developmental psychology (Pp. 219-242). New York: Wiley.

Hiroto, D.S., & Seligman, M.E.P. (1975). Generality of learned
helplessness in man. Journal of Personality and Social Psychology, 31
(2), 311-327.

Jaycox, LH., Reivich, K.J., Gillham, J ., & Seligman, M.E.P. (1994).
Prevention of depressive symptoms in school children. Behaviour
Research and Therapy, _32 (8), 801-816.

Kahn, J .S., Kehle, T.J., Jenson, W.R., & Clark, E. (1990).
Comparison of cognitive-behavioral, relaxation, and self-modeling
interventions for depression among middle-school students. School
Psychology Review- 19 (2), 196-2 1 1 .

Kaslow, N.J., Rehm, L.P., & Siegel, AW. (1984). Social-cognitive
and cognitive correlates of depression in children. Journal of

Abnormal Child Psychology, 12(4), 605-620.

88

Kaslow, N.J., Tannenbaum, R.L., Abramson, L.Y., Peterson, C., &
Seligman, M. E. P. (1983). Problem-solving deﬁcits and depressive

symptoms among children. Journal of Abnormal Child Psychology,
1 _(4),1 497- 502.

Kazdin, AB. (1981). Assessment techniques for childhood

depression: A critical appraisal. Journal of the Amerigan Academy of
Child Ps chia ,__2_()_, 358-375.

Kendall, P.C. (1993). Cognitive-behavioral therapies with
youth: Guiding theory, current status, and emerging developments.

Journal of Consulting and Clinical chhology, 61 , 235-247.

Kovacs, M. (1985). The Children's Depression Inventory (CDI).
Psychopharmacology Review. 21, 995-1 1 24.

Iaurer, R.E., Giordani, B., Boivin, M.J., & Halle, N. (1994). Effects
of depression on memory performance and metamemory in children.

Journal of the American Academy of Child and Adolescent
Psychiatg, 33(5), 679-685.

Lewinsohn, P.M., & Clarke, G.N. (1984). The coping with

depression course adolescent version: Instructor's manual for gent
course. Eugene, OR: Castalia Publishing Company.

Lewinsohn, P.M., Steinmetz, J .1.., Iarson, D.W., & Franklin, J.
(1981). Depression-related cognitions: Antecedent or consequence?

Journal of Abnormal Psychology, _9Q(3), 213-219.

McCauley, E., Mitchell, J .R., Burke, P., & Moss, S. (1988).
Cognitive attributes of depression in children and adolescents.

Journal of Consulting and Clinical Psychology, 56(6), 903-908.

Mischel, W., Shoda, Y., & Rodriguez, M1. (1989). Delay of
gratiﬁcation in children. Science 244 (4907), 933-938.

 

Nolen-Hoeksema, S., Girgus, J .S., & Seligman, M.E.P. (1986).
learned helplessness in children: A longitudinal study of depression,

achievement, and explanatory style. Journal of Personality and
Social Psychology, 51 (2), 435-442.

89

Nolen-Hoeksema, S., Girgus, J .S., & Seligman, M.E.P. (1992).
Predictors and consequences of childhood depressive symptoms: A 5-

year longitudinal study. Journal of Abnormal Psychology, 101, 405-
422.

Offord, D.R., Boyle, M.H., Szatmari, P., Rae-Grant, N.I., Links, P.S.,
Cadman, D.T., Byles, J .A., Crawford, J.W., Munroe Blum, H., Byrne, C.,
Thomas, H., Woodward, CA. (1987). Ontario child health study.

Archives of General Psychiatg, 44, 83 2-836.

Okolo, OM. (1992). The effects of computer-based attribution
retraining on the attributions, persistence, and mathematics
computation of students with learning disabilities. Journal of
Learning Disabilities,_2_5_(5), 327-334.

Peterson, C.P., & Bossio, L.M. (1991). Health and optimism.
New York: The Free Press.

Peterson, C., Seligman, M., & Vaillant, G. (1988). Pessimistic
explanatory style is a risk factor for physical illness: A thirty-ﬁve-

year longitudinal study. Journal of Personalig and Social
Psychology, _5_5_, 23-27.

Rizley, R. (1978). Depression and distortion in the attribution
of causality. Journal of Abnormal Psychology, _8_7_ (1 ), 32-48.

Robin, A.L., & Foster, S.L. (1989). Negotiating parent-adolescent

conﬂict: A behavioral-family systems approach. New York: The
Guilford Press.

Rutter, M. (1988). Epidemiological approaches to
developmental psychopathology. Archives of General Psychiatty,_4_5_,
486-495 .

Saylor, C.F., Finch, A.J., Spirito, A., & Bennett, B. (1984). The
Children's Depression Inventory. Journal of Consulting and Clinical
Psychology, _5_2_ , 95 5-967.

Seligman, M.E.P. (1991). Learned optimism. New York: Alfred
A. Knopf.

9O

Seligman, M.E.P., Castellon, C., Cacciola, J ., Schulman, P.,
Luborsky, I.., Ollove, M., & Downing, R. (1988). Explanatory style
change during cognitive therapy for unipolar depression. Journal of

Abnormal Psychology, 97, 1 3-1 8.

Seligman, M.E.P., Peterson, C., Kaslow, N.J., Tannenbaum, R.L.,
Alloy, L.B., & Abramson, LY. (1984). Attributional style and
depressive symptoms among children. Journal of Abnormal

Psychology, 83, 235-238.

Sherer, M., & Adams, CH. (1983). Construct validation of the
Self-efﬁcacy scale. Pachological Reports, _53, 899-902.

Sherer, M., Maddux, J .E., Mercandante, B., Prentice-Dunn, S.,
Jacobs, B., & Rogers, RW. (1982). The Self-efﬁcacy Scale:
Construction and validation. Psychological Remnaﬂ, 663-67 1.

Wachtel, EF. (1994). Treating troubled children and their
families. New York: Guilford Press.

Ward, L.G., Friedlander, M.L., Silverman, W.K. (1987).
Children's depressive symptoms, negative self-statements, and causal
attributions for success and failure. Cognitive Therapy and Research,
_1_1_(2), 215-227.

Zarantonello, M., Slaymaker, F., Johnson, J., & Petzel, T. (1984).
Effects of anxiety and depression on anagram performance, ratings
on cognitive interference, and the negative subjective evaluation of

performance. Journal of Clinical Psychology, _4_0_( 1 ), 20-25.

Zimmerman, MA. (1990). Toward a theory of learned
hopefulness: A structural model analysis of participation and

empowerment. Journal of Research in Personally, 14, 71-86.

APPENDICES

APPENDIX A:
SOURCES OF CHILD MEASURES

9 1
Sources of Child Measures

Children's De ression Invento CDI
Kovacs, M. (1985). The Children's Depression Inventory (CDI).
Psychopharmacology Review. 2_1__, 995-1 1 24.

Perceived Competence Scale for Children
Harter, S. (1982). The Perceived Competence Scale for Children.

Child Development, 53, 87-97.

Self-efﬁcag Scale
Sherer, M., Maddux, J .E., Mercandante, B., Prentice-Dunn, S.,

Jacobs, B, & Rogers, R.W. (1982). The Self-efﬁcacy Scale: Construction
and validation. Psychological Reports, 5 1, 66 3-67 1.

Children's Attributional Sgle Questionnaire (CASQ)
Seligman, M.E.P, Kaslow, N.J., Alloy, LB, Peterson, C.,

Tannenbaum, R., & Abramson, L.Y. (1984). Attributional style and
depressive symptoms among children. Journal of Abnormal

Psychology, 93, 235-238.

The anagram task was developed by the present author and is
provided in Appendix D.

The Cross-out task was developed by the present author.

APPENDIX B:
SOURCES OF ADULT MEASURES

92

Sources of Adult Measures

Child Behavior Checklist for Ages 4 - 18 (CBCL)
Achenbach, T.M., & Edelbrock, CS. (1983). Manual for the Child

Behavior Checklist and Revised Child Behavior Proﬁle. Burlington,
University of Vermont.

The Parent Questionnaire was developed by the present author and
is provided in Appendix E.

APPENDIX C:
HOMEWORK ASSIGNMENTS

93

HOMEWORK ASSIGNMENTS FOR THE EXPLANATORY STYLE
GROUP

Week 1: "Write about two situations in which you felt sad, angry, or
embarrassed after something bad happened. What was the belief
that led you to feel that way? Give two examples of your own and
don't forget to write down the belief that goes with the feeling, Don't
try to write down the worst thing that has ever happened to you;
just give a couple examples from today or yesterday when things
didn't happen the way you hoped or expected they woul ."
(Examples provided.)

Week 2: "Complete ﬁve Adversity-Belief-Consequence chains."
(Example provided.)

Week 3: "Use one of your A-B-C chains where you didn't like the
consequence of a belief. Write down ﬁve other possible beliefs about
the situation. Use evidence to dispute the original belief."

Week 4: "Make up two examples of A-B-C-D-E chains." (Example
provided.)

HOMEWORK ASSIGNMENTS FOR THE RELAXATION TRAINING
GROUP

Week 1: "Write about two times when you felt tense or stressed.
Then tell how you got yourself to relax." (Examples provided.)

Week 2: "Before our next meeting, tense and relax the muscles that
we went through in our last session. For each one, tense for 2
seconds and relax for 10 seconds. Then relax each one for 10
seconds without tensing beforehand. Also, think of one situation
where you would like to be able to relax more, and write it down for
next time." (Exercises listed.)

94

Week 3: "For the next session, practice each of the following
exercises. Do each exercise for only a few seconds at a time."
(Exercises described.)

Week 4: "Write down two situations where you could use the
relaxation techniques that we learned in group sessions."

APPENDIX D:
ANAGRAMS

95

ANAGRAMS

Start Time:
Finish Time:

 

 

Your Name:

 

Anagrams are groups of letters that you can re-arrange to make a word. Try to
solve these 10 anagrams by re-arranging the letters to make words. Some of
the anagrams are more difﬁcult than others. Most people won't be able to get
all the correct answers. Do your best and take as much time as you want.

Example: T R P I N Answer: PRINT

1. RPPAE

2. LWCON

3. SBOKO

4. OLROC

5. NURDE

6. UOHES

7. OR'ITI

8. GIHLT

9. SYNAN

1 O. ELRYA

96

ANAGRAMS

Start Time:

 

Finish Time:

 

Your Name:

 

Anagrams are groups of letters that you can re-arrange to make a word. Try to
solve these 10 anagrams by re-arranging the letters to make words. Some of
the anagrams are more difﬁcult than others. Most people won't be able to get
all the correct answers. Do your best and take as much time as you want.

Example: T R P I N Answer: PRINT

1. SERPS
2. UIMCS
3. EXOSB
4. ORDAI
5. CIPHT
6. RAHST
7. MSGUE
8. PALEP
9. RZEIH

10. SLMAL

97

ANAGRAMS

Start Time:
Finish Time:

 

 

Your Name:

 

Anagrams are groups of letters that you can re-arrange to make a word. Try to
solve these 10 anagrams by re-arranging the letters to make words. Some of
the anagrams are more difﬁcult than others. Most people won't be able to get
all the correct answers. Do your best and take as much time as you want.

Example: T R P I N Answer: PRINT

1. HARIC
2. PTNIA
3. SMGAE
4. TFURI
5. MTOHN
6 SDERS
7 HATYN
8 NREGE
9. FINAC

10. PSNAT

98

ANAGRAMS

Start Time:

 

Finish Time:

 

Your Name:

 

Anagrams are groups of letters that you can re-arrange to make a word. Try to
solve these 10 anagrams by re-arranging the letters to make words. Some of
the anagrams are more difﬁcult than others. Most people won't be able to get
all the correct answers. Do your best and take as much time as you want.

Example: T R P I N Answer: PRINT

1. BALTE
2. CDAYN
3 (XIICU
4. BSDRI
5. TWEIH
6. ROSET
7 NRETE
8. SLSAC
9. REGNA
10. UPCNH

99

ANAGRAMS

Start Time:
Finish Time:

 

 

Your Name:

 

Anagrams are groups of letters that you can re-arrange to make a word. Try to
solve these 10 anagrams by re-arranging the letters to make words. Some of
the anagrams are more difﬁcult than others. Most people won't be able to get
all the correct answers. Do your best and take as much time as you want.

Example: T R P I N Answer: PRINT

1. KISCT
2. HTGLI
3. CRAHT
4. HUSOT
5. LAGNE
6. HRITS
7. PAYKP
8. SRAGS
9. ROKYS

10. ZPREI

APPENDIX E:
PARENT QUESTIONNAIRE

100

Parent Questionnaire

 

Your Child's Name: Child's Age:
Child's Grade Level:
Your Name: Your Age:

 

1. What is your relation to this child? (Circle one):

Mother Father Stepmother Stepfather Grandmother

Grandfather Aunt Uncle Other

2. How many children are currently living in your home?

boys (please give ages of boys):

 

 

girls (please give ages of girls):

 

 

3. What is your ethnicity? (Check one):
African American Asian American Caucasian
Hispanic Native American
Other (Explain):

 

4. What is your child's ethnicity? (Check one):
African American Asian American Caucasian
Hispanic Native American
Other (Explain):

 

5. What is your level of education? (Check one):
Finished junior high school
Some college

Completed a degree from a trade school or vocational program
Completed a 2-year college degree
Completed a 4-year college degree

Finished high school

 

 

 

6. What is your total yearly household income? (Check one):

Less than $10,000 $10,000 - $20,000
$20,000 - $35,000 $35,000 - $50,000
$50,000 - $70,000 Greater than $70,000

7. What is the marital status of the child's biological parents?

Married Divorced Separated

 

Completed a graduate degree

Widowed Single (the biological parents were never married

to each other)
Other (Explain):

 

101

8. How is your child doing in school now? (Check one):
Excellent (A average) Satisfactory (B or C average)
Below Average (D average) Failing (F average)

The following questions ask how much time your child spends in
various activities. Please be as accurate as you can about your
child's behavior.

9a. How long does your child study each week?

9b. Does your child study every day? (Circle one): Yes No

9c. How many minutes does your child study at one time? minutes

9d. When your child stops studying, do you usually encourage him or her to
continue to study for a while longer?
(Check one): Yes (If YES, please go to 9e)
No (If NO, please go to 9f)
9e. If YES, does your child usually agree to continue to study when you ask
him or her to do so? (Circle one): Yes No

9f. How often does your child "give up" when studying, and refuse to complete
an assignment?

(Circle one): Almost never Once a month Once a week Almost every day

10. Does your child participate in a sport? (Circle one): Yes No
If NO, please go to Question 1 1.
If YES, How long has your child played this particular sport?

10a. How much time does your child practice the sport each week?
10b. Does your child practice the sport every day? (Circle one): Yes No

10c. How many minutes does your child practice the sport at one time?
minutes

102

10d. When your child stops practicing the sport, do you usually encourage him
or her to continue to practice for a while longer?
(Check one): Yes (If YES, please go to 10e)

No (If NO, please go to 101’)

 

10e. Does your child usually agree to continue to practice the sport when you
ask him or her to do so?

(Circle one): Yes No

10f. How often does your child "give up" when practicing the sport, and refuse
to continue to play?

(Circle one): Almost never Once a month Once a week Almost every day

11. Does your child play a musical instrument? (Circle one): Yes No
If NO, please go to Question 12.
If YES, How long has your child played this particular instrument?

11a. How much time does your child play the instrument each week?

minutes

1 1b. Does your child play the instrument every day? (Circle one): Yes No

11c. How many minutes does your child play the instrument at one time?

minutes
11d. When your child stops practicing, do you usually encourage him or her to
continue to play the instrument for a while longer?
(Check one): Yes (If YES, please go to Me)
No (If NO, please go to 11f)

103

1 1e. If YE S, does your child usually agree to continue to play the instrument
when you ask him or her to do so?

(Circle one): Yes No

11f. How often does your child "give up" when practicing, and refuse to
continue to play the instrument?

(Circle one): Almost never Once a month Once a week Almost every day

12. Is your child involved in any group or activity (besides sports or music
lessons) outside of school? (Check one):
No (If NO, please stop here.)
Yes (If YES, What is the activity? )

12a. How long has your child participated in this activity?
months

 

12b. How much time does your child participate in the activity each week?
minutes
12c. Does your child participate in the activity every day?
(Circle one): Yes No
12d. How many minutes does your child participate in the activity at one

time?
minutes

He. How often does your child refuse to attend activities or group meetings?

(Circle one): Almost never Once a month Once a week Almost every day

 

 

N

1

2

"1

   

1111111