. .

. . ., .

My!"
1' ht

." <

gm

«‘1 .1 «4.
u: an. .... .
.\.. yam "3
Quad.» .wz.’
var.» u. v
5‘ -

5‘?!

gym

4 .ulmv s.
mu u v
‘ .-

395?.

‘.‘ . ’ .
u .._.-~

. .v.v.r.r
.h ..'<.'u~.ulln
< x. “tut-1‘

29":

F

.\
15‘4“
. . ‘

wsssisei
5 ‘24:?
‘EW '
8
519;

4‘23“)
”NV 1:. _ . V -
, - .--.-\A-t

(”WI
‘5?

a.

pr
I I
’v'rt‘tfi

«v

Ag}

.‘Pr’?
,. .

4..
:"u' :t

. , ”-4. .
‘ 5. Au: an-
. J.“ .‘xuv ,., . .._
.. ‘-\ I.’a'.\-L

u »

 

 

r ‘ 'r
fib'a‘fl‘i

Wash-n:

3?

 

mm

”3".

‘

Eifiéismr

nui‘w‘

"Wu-m

'u-r-

x'A‘

Innu-

1......
4. ,, .
~ .- 9

4

.. ,w

 

. 3....
' 3'71 1" ' L'
,' l‘ .g ..

..,...§...‘.v,...rv
,,._,‘ ‘,A.,...._..
“0,.“ ...-.
... 4. .
. h” ‘ .».“~
.,,,,.‘V,.. ..,

r

,m

‘7 r».
WY)... "
. ......
,0.

l

I
— n .
n-4,.

~~n¢
" “fluff... .4
.-.

_.,..
av-
. .wm.‘
I 5 4- ‘h‘

r 'f

- «- .,.,‘. ,1 I" r I
rw-g' - 1* ' .. . or ~qu— "’J"""' .

, o u -3‘ “ N "F1, ’ .ZIJL-J.’ .‘2,.':.'.'! ‘J.’.’..

Jit'f"; 3".“3 ";’.’."£\" ‘21“: ‘ . . WMMVJV¢?JI.'M~

 

 

.' l (““3
y ht-Jvth

lulll'qlllllllml lilllllll

00898 9331

This is to certify that the

thesis entitled

Toward Validating the Diathesis-Stress
and Causal Mediation Components .
of the Hopelessness Theory of Depre331on

presented by

Mark Hudson Wagner

has been accepted towards fulfillment
of the requirements for

Master of Arts degree in Psychology

 

 

gaziiiEQZZe~ci&:::f

Major professor

May 16, 1991
ac

 

MS U is an Affirmative Action/Equal Opportunity Instirution

 

 

LlfifiARY
Michigan State
l University

 

 

 

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

DATE DUE DATE DUE DATE DUE

l:]__JI-
WET—l

HQE'
*7 —lfl

MSU Is An Affirmative Action/Equal Opportunlty Institution
cmmt

      

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

TC

TOWARD VALIDATING THE DIATHESIS-STRESS AND CAUSAL MEDIATION
COMPONENTS OF THE HOPELESSNESS THEORY OF DEPRESSION

BY

Mark Hudson Wagner

A THESIS

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

MASTER OF ARTS

Department of Psychology

1991

r
1
a,
O
.
0v
.-
4
i
u

i -‘
0?
-
-
C
.1
4-.
’l
,.
.4;
.,
1;
A
.-
'
a
r,

 

TC

.,. ’7
/‘

I

J.—v
I

4354:.

ABSTRACT ,

TOWARD VALIDATING THE DIATHESlS-STRESS AND CAUSAL MEDIATION
COMPONENTS OF THE HOPELESSNESS THEORY OF DEPRESSION

BY

Mark Hudson Wagner

The Diathesis-Stress and Causal Mediation components of the
Hopelessness Theory of Depression (HTD; e.g., Abramson, Metalsky and Alloy,
1988) were evaluated by measuring 43 undergraduate students’ affective
responses to midterm exam grades. In an attempt to logically isolate the
depressive responses proposed by HTD (i.e., the etiologically defined
hopelessness subtype), each subject’s baseline condition regarding attributional
style, recent important negative life events (INLEs), and depression was evaluated.
While 17 subjects viewed their exam outcome as negative, no pattern of enduring
depressive mood response or feelings of hopelessness was shown, and therefore,
hypotheses regarding the prediction of such responses were not viewed as
adequately tested. Further, the consistency between finding no enduring
depressive mood response in a sample from which depressives were selected and
trait theory for depression was discussed. Issues regarding the measurement of

INLEs, and methodological and theoretical implications are addressed.

For my parents, Charles and Elizabeth Wagner,
who have taught me very much.

Ul'l
me

ACKNOWLEDGMENTS

Special thanks to Dr. Dozier W. Thornton, the chairman of my committee, for
his constant support and patience through this project. Thanks also to the members
of my committee: Dr. Mark E. Rilling, Dr. Michael K. Lindell, and Dr. John E.
Hunter. Thanks also to Dr. Leonard VanderJagt for his contributions during the
development of this project, and his support throughout. I would like to recognize
further Dr. Thornton and Dr. Rilling for their special inspiration, Dr. Thornton and
Dr. Lindell for their many valuable consultations, and Dr. Hunter for his direction
on addressing the Trait Theory of Depression.

Thanks to Dr. Lindell and to Leslie Wolowitz for allowing me to collect data
in their classes. Thanks to Brenda Mayne for performing the subject solicitation
speeches and for her help in collecting data. Thanks to Kent Phillippe for his help
with SPSS. Thanks also to Denise Bachman and Dave Herrod for their help in the
questionnaire preparation and data collection and organization, and more thanks to
Denise for her help in a hundred or more hours of data entry.

Thanks to Maureen Marks, David Rosenberg, and Alexis Vlahos for their
special friendship and support.

Love and very special thanks to Christine for her love, friendship, support,
understanding, encouragement, guidance, effort and sacrifice, and for growing with
me in our marriage, all while pursuing her doctorate. I am truly fortunate.

iv

 

 

Hav
I we

On

Having taken five years to complete this thesis,
I would like to share some thoughts --

On Procrastination:

A boy who did always procrastinate

When asked why all of his poems were late
Said it took too much time

To find words that would rhyme

- A ”Limerick“ by the author written to complete an overdue
poetry assignment for a Creative Writing class in the 9th grade.

How many graduate students does it take to screw in a light bulb?
One -- but it takes him nine years.

- from MW William Novak and
Moshe Waldoks (Eds)

My mother said, "You won ’t amount to anything because you procrastinate!"
I said, 'Ilust wait. "

-- Judy Tcnuta. quoted in W

Argue for your limitations, and sure enough, they 're yours.
- from m by Richard Bach

535’?

3'

TABLE OF CONTENTS

INTRODUCTION
The Learned Helplessness Theory of Depression (LHTD)

The Reformulated Learned Helplessness Theory of
Depression (LHTD-R)

The Hopelessness Theory of Depression (HTD)

Toward an Appropriate Test of the Hopelessness Theory of
Depression (HTD)

The Diathesis-Stress Component of the Hopelessness
Theory of Depression

The Causal Mediation Component of the Hopelessness
Theory of Depression

Investigation of the Diathesis-Stress and Causal
Mediation Components of the Hopelessness Theory of
Depression (HTD)

Critical Appraisals of the Reformulated Learned
Helplessness Theory of Depression (LHTD-R) and the
Hopelessness Theory of Depression (HTD)

A More Appropriate Test of the Hopelessness Theory of
Depression (HTD)

Diathesis-Stress Component Hypotheses

Causal Mediation Component Hypotheses

vi

10

18

23

24

42

43

METHOD

Subjects

Measures

Procedure

Statistical method

RESULTS

Tests of Assumptions Necessary for an Analysis of Partial Variance
Tests of Hypotheses: Description of Statistics

Tests of the Diathesis-Stress Hypotheses

Tests of the Causal Mediation Hypotheses

Descriptive Statistics

DISCUSSION

FOOTNOTES

APPENDIX A: Attributional Style Questionnaire
APPENDIX 8: Particular Attribution Questionnaire
APPENDIX C: Grade Aspirations Questionnaire
APPENDIX D: Multiple Affect Adjective Check List-Revised, Today Form
APPENDIX E: Beck Depression Inventory

APPENDIX F: Beck Hopelessness Scale

APPENDIX G: Life Events Inventory

APPENDIX H: Solicitation Speech for “The Mood Study"

APPENDIX I: Solicitation Speech for the “Personality Style“ Study

vii

 

45
45
46
48
52
57
57
59
63
69
76
81
102
103
121
125
127
129
131
133

137

APPENDIX J: Subject Debriefing Information Handout 139

APPENDIX K: Tests of Homogeneity of Regression for Analyses of
Partial Variance 141

APPENDIX L: Correlation Matrices of All Independent and Dependent
Variables for Various Portions of the Full
Subject Sample 142

LIST OF REFERENCES 152

viii

 

'rilfl

Ta

Table 1:

Table 2:

Table 3:

Table 4:

Table 5:

Table 6:

Table 7:

Table 8:

Table 9:

Table 10:

Table 11:

LIST OF TABLES

Summary of Results for Test of Diathesis-Stress
Hypothesis 1

Summary of Results for Test of Time 2 Comparison
to Diathesis-Stress Hypothesis 1

Summary of Results for Test of Diathesis-Stress
Hypothesis 2

Summary of Results for Test of Time 2 Comparison
to Diathesis-Stress Hypothesis 2

Summary of Results for Test of Diathesis-Stress
Hypothesis 3

Summary of Results for Test of Time 2 Comparison
to Diathesis-Stress Hypothesis 3

Summary of Results for Test of Causal Mediation
Hypothesis 1

Summary of Results for Test of Positive Exam Outcome
Comparison to Causal Mediation Hypothesis 1

Summary of Results for Test of Total Sample Comparison
to Causal Mediation Hypothesis 1

Summary of Results for Test of Causal Mediation
Hypothesis 2

Summary of Results for Test of Time 2 Comparison to
Causal Mediation Hypothesis 2

ix

66

70

7O

70

72

72

Table 12: Summary of Results for Test of Causal Mediation

Hypothesis 3 74
Table 13: Summary of Results for Test of Time 2 Comparison to

Causal Mediation Hypothesis 3 74
Table 14: Summary of Results for Test of Causal Mediation

Hypothesis 4 75
Table 15: Summary of Results for Test of Time 2 Comparison to

Causal Mediation Hypothesis 4 75
Table 16: Summary of Results for Test of Causal Mediation

Hypothesis 5 77
Table 17: Summary of Results for Test of Time 2 Comparison to

Causal Mediation Hypothesis 5 77
Table 18: Summary of Descriptive Statistics 79

Table 19: Summary of Results of the Metalsky et al. (1987) Study
that Correspond to the Test of the Time 2 Comparison
to Diathesis-Stress Hypothesis 1 87

Table 20: Summary of Results of the Test of the Time 2
Comparison to Diathesis-Stress Hypothesis 1 that
Correspond to the Metalsky et al. (1987) Results in
Table 19 87

Table 21: Number of Students who Achieved each Level of Grade
on their Exam 93

Table A: Summary of Tests for Homogeneity of Regression for
Analyses of Partial Variance 141

Table 8: Correlation Matrix of All Variables for Subjects With
Complete Data, 3 Low Beck Depression Inventory Score,
and No Recent Important Negative Life Events 142

Table C: Correlation Matrix of All Variables for Subjects Wlth
Complete Data, 3 Low Beck Depression Inventory Score,
No Recent Important Negative Life Events, and Who Had
a Negative Exam Outcome 143

Table

Table
Table
Table

Table

Table

Table

Table

: Correlation Matrix of All Variables for Subjects With

Complete Data, a Low Beck Depression Inventory Score,
No Recent Important Negative Life Events, and Who Had
a Positive Exam Outcome

: Correlation Matrix of All Variables for Subjects With

Complete Data

: Correlation Matrix of All Variables for Subjects With

Complete Data, and Who Had a Negative Exam Outcome

: Correlation Matrix of All Variables for Subjects With

Complete Data, and Who Had a Positive Exam Outcome

: Correlation Matrix of All Variables for Subjects With

Complete Data and a Low Beck Depression Inventory
Score

Correlation Matrix of All Variables for Subjects With

Complete Data and a High Beck Depression Inventory
Score

: Correlation Matrix of All Variables for Subjects With

Complete Data and No Recent Important Negative Life
Events

: Correlation Matrix of All Variables for Subjects With

Complete Data and a Recent Important Negative Life
Events

144

145

146

147

148

149

150

151

 

 

figure 1:

Figure 2:

Figure 3:

figure 4:

Figure 5:

Figure 6:

figure 7:
figure 8:

figure 9:

LIST OF FIGURES

The Cognitive Context and Expected Consequences of

Six Types of Causal Attributions

Examples of Eight "Types" of Causal Attributions for
Positive or Negative Events

Defining Characteristics of Depressogenic and
Depression Resistant Attributional Styles

Defining Characteristics of Five Types of Causes

The Etiological Chain Proposed by the Hopelessness
Theory of Depression

Possible Explanations for the Presence or Absence of
Depressive Symptoms

The Complete Study as Conducted
'THE MOOD STUDY“

The “Personality and Style“ Study

xii

13

14

19

42

42

42

ASO

BDI

 

BHS

INLE

LEI

 

MAAi
MAM

PAQ

 

ASQ
BDI

BHS
HTD

INLE

LEI

LHTD-R

MAAC L
MAACL-R

PAG

ABBREVIATIONS

Attributional Style Questionnaire.

Beck Depression Inventory.

Beck Hopelessness Scale.

The Hopelessness Theory of Depression.

Important Negative Life Event, (and the absence of an
important positive life event).

Life Events Inventory.
The Learned Helplessness Theory of Depression.

The Learned Helplessness Theory of Depression -
Reformulated.

Multiple Affect Adjective Check List, Today Form.
Multiple Affect Adjective Check List - Revised, Today Form.

Particular Attribution Questionnaire.

xiii

INTRODUCTION

The research reported here was designed to evaluate the Hopelessness
Theory of Depression (HTD; Abramson, Metalsky and Alloy, 1986; Abramson,
Metalsky and Alloy, 1988; Alloy, Abramson, Metalsky and Hartlage, 1988). HTD
consists of two fundamental theoretical components, the Diathesis-Stress
component and Causal Mediation component. A recent test of HTD (Metalsky,
Halberstadt and Abramson, 1987) suggested support for both components of
the theory. Though the study of Metalsky et al. (1987) showed a great
improvement in theoretical soundness over an earlier test of HTD (Metalsky,
Abramson, Seligman, Semmel and Peterson, 1982), their test used a
methodology that was inconsistent with some of the basic tenants of HTD.
These inconsistencies leave their results open to wide interpretation. Therefore,
by adapting the methodology of Metalsky et al. (1987) to be consistent with
HT D, this study was designed to provide a more sound and more interpretable
test of HTD.

Before discussing the details of this study, some background information
relevant to its development will be presented. first, the evolution of HTD from its

roots in Seligman's (1975) Learned Helplessness Theory of Depression (LHTD)

2

will be traced. Next, the research implications of HTD will be presented. Then a
discussion of research testing the Diathesis-Stress and Causal Mediation
components of HTD will be presented. Included will be a summary and a
critique of the Metalsky et al. (1987) study. Then a discussion of how the
shortcomings of the Metalsky et al. (1987) study were addressed in the present,

and finally, the resulting research will be reported.

Tern Hll n Th fDrinLHTD

The first reports of the learned helplessness phenomenon came from
experiments noting that dogs exposed to inescapable electrical shock later
showed motivational, cognitive, and emotional deficits (Overmier and Seligman,
1967; Seligman and Maier, 1967). The deficits were assessed in later trials when
the contingencies had changed such that escape from the shock was then
possible. The fact that the dogs initiated very few attempts to escape the shock
was seen to evidence their motivational deficit. When a dog did successfully
escape the shock, it rarely followed that experience with another escape
attempt in the next trial. This apparent failure to learn from the successful
escape experience was seen to evidence the dogs” cognitive deficit. Their
emotional deficit was evidenced by their not appearing to demonstrate much
overt emotionality while they were being shocked in these later trials.

Seligman (1975) noted that depressed humans show the same three
classes of deficits. Depressed humans show a decline in their initiation and

persistence of voluntary responses (motivational deficit), in their perception of

3

opportunities and controllable situations (cognitive deficit), and show greater
levels of sadness with lowered self-esteem (emotional deficit). Given the
similarities between the deficits demonstrated by dogs exposed to inescapable
shock and those demonstrated by depressed humans, Seligman (1975)
suggested that the cause of human depression may be similar to the cause of
the dog’s deficits, and proposed that depressed humans have expectations that

they cannot control their outcomes.

1: ' -ofrm l :- L:.rn:- H pl: n Thu cf Dar: l-n _ I-R
Abramson, Seligman and Teasdale (1978) noted that LHTD did not

explain certain qualitative variations in both how people perceive the
uncontrollability of their situations and the patterns of deficits they demonstrate
as a result of those perceptions (i.e., experiencing personal vs. universal
helplessness,‘ and the generality and chronicity of their helplessness deficits).
To account for these variations, Abramson et al. (1978) proposed a
reformulation (LHTD-R) of LHTD that emphasized the importance of the causal
attributions people make to explain their helpless situations. They suggested
that causal attributions for positive or negative outcomes could be meaningfully
described along three orthogonal, continuous dimensions Ge, internal vs.
external, stable vs. unstable, and global vs. specific dimensions of causal
attributions). Figure 1 presents the cognitive context of each type of causal
attribution (where "types" are the endpoints of each of the proposed dimensions

of causal attributions) and also presents the consequences people are

 

 

 

 

 

 

 

 

The cognitive context of each type of causal attribution The corresponding
consequences expected

The individual's Type of causal A corresponding for the individual if
belief regarding the attribution generic example of his/her view is that

nature of his/her made for the each type of their situation is
situation situation causal attribution important and helpless

Relevant others Attribute the I am somehow the 'Personal helplessness,
(i.e., peers) in my situation to cause of this sadness. and
situation would view INTERNAL causes situation lowered self-esteem

it as controllable (DAS)

Relevant others Attribute the Something or someone Universal helplessness.
(i.e., peers) in my situation to else is the cause sadness, and no
situation would view EXTERNAL causes of this situation deficit in self-esteem

it as not controllable (RAS)
Hy situation is due to Attribute the This situation was Demonstration of
factors that impact situation to caused by something deficits in a wide
many areas in my life GLOBAL causes which will effect my variety of areas
(DAS) life in many ways
Hy situation is due to Attribute the This situation was Demonstration of
factors relevant only situation to caused by something specific deficits
to this particular SPECIFIC causes which will not effect related to the
situation (RAS) my life in many ways situation at hand
My situation is due to Attribute the This situation was Deficits are
longstanding factors situation to caused by something demonstrated in a
likely to persist as STABLE causes which will continue chronic pattern
factors in my life (DAS) to impact my life
in the future
My situation is due to Attribute the This situation was Deficits are
transient factors that situation to caused by something demonstrated in a
are neither persistent DISTAILI causes which will not transient fashion
nor unalterable (RAS) continue to effect my
life in the future

 

 

 

 

 

 

Tigure 1
WWW

uggg. DAS - Depressogenic Attributional Style, and RAS - Depression Resistant
Attributional Style.

' - Abramson et al. (1978) address the emotional deficit described by Seligman (1975) by
distinguishing self-esteem and affective deficits. They contend that effective deficits
in depression result from the expectation that bad outcomes which are important to the
individual will occur, or that good outcomes which are important to the individual will
not occur, independent of the controlability of these outcomes. Further. they state that
“depressed individuals who believe their helplessness is personal show lower self-esteem
than individuals who believe their helplessness is universal” (p. 66)

\h

E

 

We.

AV

 

LIL

 

5

expected to experience for making each type of causal attribution regarding a
situation they perceive as helpless. Figure 2 (adapted from Abramson et al.,
1978, p.57) demonstrates how the three attributional dimensions can fully
interact to qualitatively describe and distinguish various attributions regarding
the cause of a single event, whether positive or negative. Furthermore, Figure 2
demonstrates the eight (2 x 2 x 2 = 8) unique combinations of attributional
"types" that might be represented in causal attributions for a single event. Of
course, because the attributional dimensions are continuous, causal attributions
may be represented anywhere along each of the three dimensions, thus
yielding an infinite variety of combinations of attributional characteristics possible
between these eight combinations. Also, two different attributions for a single
event might be represented similarly across the attributional dimensions. For
example, “My conversation sometimes interests him/her“ and "I sometimes
dress in a way that he/she finds attractive" are both internal, specific, unstable
causal attributions that might be made for a successful interpersonal encounter.

Further examination of Figure 1 reveals that the combined consequences
expected for attributing helplessness to causes which are m, global and
stable are consistent with depressive symptomatology (i.e., lowered self-esteem,
sadness and a broad pattern of cognitive and motivational deficits that are
chronic in nature). For this reason, such an attribution has been hypothesized
to be W, and the tendency to make such attributions has been

referred to as a h th i r ni ri i n I l (e.g., Abramson
et al., 1986; Abramson et al., 1988; Alloy et al., 1988). Correspondingly,

 

 

 

 

 

 

 

 

 

 

 

 

 

 

S T A B L E U N S T A B L E
I P
N O I am attractive My conversation sometimes
g S to men/women interests men/women
G R
N
L A N I am unattractive Hy conversation sometimes
0 L g to men/women bores men/women
B E P People are accepting
X 0 of others for potential People sometimes get
A E 8 future relationships in friendly moods
L a '
N People are competitive
A N with others in potential People sometimes get
L 5 future relationships in rejecting moods
I P
N O I am attractive My conversation sometimes
S g S to him/her interests him/her
P R
N
E A N I am unattractive Hy conversation sometimes
L E to him/her bores him/her
C G
I E P He/She is accepting
X 0 of others for potential He/She sometimes gets
P T 8 future relationships in a friendly mood
E
I R
N He/She is competitive -
C A N with others in potential He/She sometimes gets
L E future relationships in a rejecting mood
G

 

 

Figure 2
WWW“

£253. In these examples, it is assumed that the situation is one where the desired
outcome is gaining the acceptance of another. POS - Positive life event: In this
example, acceptance (i.e., a successful interpersonal encounter). NEG - Negative
life event: In this example. rejection (i.e., a failed interpersonal encounter).

7
attributing helplessness to causes that are external, epeeLfie and meg—able is

expected to result in no loss of self-esteem and contextually circumscribed
affective, cognitive and motivational deficits that are transitory in nature. The
course and nature of these consequences are contrary to depression in any
enduring, disabling sense, and represent instead limited reactions to what are
viewed as isolated events. Recognizing that LHTD-R suggests that the
moderate character of these deficits are due the nature of the causal
attributions one makes rather than to the event that elicited them (and hence
the deficits), such an attribution will be referred to as _dep_r_es_sio_n_res_iiterlt and
the tendency to make such an attribution will be referred to as a depLeseigu
William?

Abramson et al. (1978) suggested that these qualitative differences in the
patterns of causal attributions people make regarding both the positive and
negative outcomes of past and present events (i.e., depressogenic or
depression resistant attributional styles) can serve to pLediQt the recurrence of
their expectations for outcomes in future situations. However, they stressed that
it is ultimately the expectations for future outcomes that determine the
occurrence of helplessness deficits Ge, “when highly desired outcomes are
believed improbable or highly aversive outcomes are believed probable, and the .
individual expects that no response in his repertoire will change their Iikelihood;“
Abramson et al., 1978, p.68). figure 3 displays the defining characteristics of a
depressogenic and depression resistant attributional styles, and their

corresponding implications for future expectations and consequent symptoms.

 

Detroit
A":".

O‘.
.I

q

secret
icsLI
A""~

.\mmum

5:71

 

.1 ‘~s¢l
' g Stltc
“"‘ Kw. -

 

 

 

Nature of .Nature of Expectations Consequent
causal causal for outcomes symptoms
attributions attributions of future expected
for FAILURE for SUCCESS experiences
experiences experiences
Imminence of Feelings of
Depressogenic INTERNAL EXTERNAL continued helplessness,
Attributional GLOBAL SPECIFIC negative motivational and
Style and and outcomes cognitive deficits
STABLE UNSTABLE independent with sadness and
of responses low self-esteem
Possibility Feelings of
Depression EXTERNAL INTERNAL of future control, with
Resistant SPECIFIC GLOBAL positive intact cognitive
Attributional and and outcomes and motivational
Style UNSTABLE STABLE dependent functioning and

on responses intact self-esteem

 

 

 

 

 

 

 

Figure 3
WW:

'.. The role of causal attributions for positive events (e.g., success) in DAS and RAS is
not gxpligigly stated in Abramson et al. (1978). Rather. they note that 'the particular
attribution that depressed people choose for failure is probably irrationally disposed
toward global. stable and internal factors and, for success. possibly toward specific.
unstable, and external factors" (p.68). and that one indication for treatment is to
“change unrealistic attributions for success toward internal, stable. global“ factors
(p.69). Seligman. Abramson. Semmel and Von Baeyer (1979) state that 'the role of
attributions for good outcomes [in the onset of depression] seems less direct [than those
for bad outcomes]. Among the possibilities are that attributions to global, stable and
internal factors for good outcomes blunt the impact of bad outcomes. that such
attributions increase ‘ego strength.’ or that good outcomes are less remembered or valued
by depressives“ (pp 246-247).

 

It“. 53’

3* 91:11:: 5

fiifih.
a we):

aff'ib

9

In summary then, Abramson et al. (1978, p. 68) present the following as
an explicit statement of LHTD-R:

1. Depression consists of four classes of deficits: motivational,
cognitive, self-esteem, and affective.

2. When highly desired outcomes are believed improbable or
highly aversive outcomes are believed probable, and the individual
expects that no response in his repertoire will change their likelihood,
(helplessness) depression results.

3. The generality of the depressive deficits will depend on the
globality of the attribution for helplessness, the chronicity of the
depression will depend on the stability of the attribution for helplessness,
and whether self-esteem is lowered will depend on the intemality of the
attribution for helplessness.

4. The intensity of the deficits depends on the strength, or
certainty, of the expectation of uncontrollability and, in the case of

affective and self-esteem deficits, on the importance of the outcome.

 

:4”:
33’s ..

heipfeSSn-es
0.‘ their hes;
7). Second.
'Heiplessnef
'BXDECtation
aWNWom
Will. Change i
1985, p. 7; A
rei'lefining ter
Views helilies

hopeIBSSReSS

1O
ThHlnTh fDrinHTD

HTD is in small part a modification of LHTD-R, and in large part, a

painstaking clarification of LHTD-R. Both aspects of HTD are presented.

ThH In Th fDrinHTDzMifitinfth
Befermuletee Leernee Helpleeeneee Thng ef erreeeien (LHTD-B)

HTD, as its name implies, emphasizes the role of feelings of
hepeleesflese in the onset of depression (Abramson et al., 1986; Abramson et
al., 1988; Alon et al., 1988). The distinction that HTD makes between
helplessness and hopelessness is intricate. First, "helplessness" is reassigned
its original meaning from Seligman’s (1975) LHTD. Specifically, HTD views
helplessness as "the expectation that one cannot control outcomes regardless
of their hedonic valence or their likelihood of occurrence,“ (Alloy et al., 1988, p.
7). Second, “hopelessness" is assigned the meaning formerly granted
“Helplessness“ in LHTD-R. Specifically, HTD views hOpelessness as the
“expectation that highly desired outcomes are unlikely to occur or that highly
aversive outcomes are likely to occur and that no response in one's repertoire
will change the likelihood of occurrence of these outcomes,“ (Abramson et al.,
1986, p. 7; Abramson et al., 1988, p. 4; Alloy et al., 1988, p. 7). Therefore, by
redefining terms, but without changing the operational logic of LHTD-R, HTD
views helplessness as one part of the larger set of expectations that comprise

hopelessness.

 

tc irenai.
Cal-“.39 in l
INLE rathe

been); 9:,

11
By reemphasizing that hopelessness is a set of expeetajjegs, HTD

proposes that uncontrollability need not be experiegeed in order for depression
to occur. Rather, HTD proposes that depression will occur when the
amnion of uncontrollability (and the other expectations that comprise
hopelessness) are elicited. HTD suggests that a perceived occurrence of an
important negative life event (INLE) is capable of eliciting expectations of
hopelessness and a full spectrum of depressive deficits if the INLE is attributed
to internal, global and stable causes.3 Therefore, one operationally significant
change in HTD from LHTD-R is the requirement of the occurrence of only an
INLE rather than actually experiencing an uncontrollable event or situation in the
theory’s etiological model.
Another operationally significant change in HTD relative to LHTD-R is
presented by Abramson et al. (1986) when they state:
We [postulate] that attributing a negative life event to an internal cause
does not, by itself, contribute to lowering self-esteem. Instead, attributing
a negative life event to an WM cause contributes to
lowered self-esteem. This revision is based on a number of studies (e.g.,
Crocker, Alloy, & Kayne, 1987; Dweck & Licht, 1980; Janoff-Bulman,
1979) showing that internal attributions pe_r_s_e are not maladaptive and,
in some cases, may be very adaptive (e.g., attributing failure to lack of

effort leads to increased trying).4 (p. 25)

 

.14”
A“
I 5

Ear...
.s:‘:

I
\"eh
~ 0
§

is»;
ca
pa yue
a1
Hf"
g C

andrh
rheMJp-i
mrlrfiw
be 6 fl CC."
Tl“ M
a %:pm
“is in a >
n

BW€“y

12
ThHln Th fDrin TD'lrifitin fth

rlt m Hll n Th f r inLHT-R

With the exception of the theoretical modifications just mentioned, HTD is
logically identical to LHTD-R. However, the primary value of HTD is the
painstaking clarity with which it explains its logic, limits, etiological postulates
and methodological implications.

In considering the etiology of depression, HTD puts considerable weight
on the distinctions between necessary, sufficient and contributory causes, and
between proximal and distal causes. Figure 4 defines each of these types of
causes in a manner consistent with their use in HTD.

Briefly stated, HTD proposes that among the many possible subtypes of
the heterogeneous disorder of depression (Depue and Monroe, 1978), the
W is the result of a chain of contributory
causes that culminate in hopelessness, which is viewed as a proximal sufficient
cause of depression (i.e., hopelessness depression). The framework for this
etiological chain of hopelessness depression is presented in Figure 5, and will
now be explained component by component, starting with the cause most
proximal to the manifestation of hopelessness depression, working back to its
most distal contributory causes.

HTD proposes that hopelessness is a WWW
gemeeejeg. The primary implication of this is the proposition that for ell
instances in which one experiences the set of expectations by which HTD

defines hopelessness, one will then experience depression. However, HTD also

 

Secessa

 

Pretzel

\- .
alltg.

 

E
b
e

P .
. :h. ‘L...
‘30:: slaw

’U

(ID

'7!
'1

13

 

Defining probability
Type of cause statements for each Defining characteristics for each type of cause
type of cause

 

 

 

P(E/S) - 1.00 For every instance in which the SYMPTOMS have
occurred, the EVENT has always occurred first
Necessary
P(not E/S) - 0.00 The SYMPTOMS never occur when the EVENT has
not occurred first
P(S/E) - 1.00 For every instance in which the EVENT occurs,
the SYMPTOMS will then also occur
Sufficient
P(E/not S) - 0.00 If the SYMPTOMS have not occurred, then the
EVENT must not have occurred either
P(S/E) > P(S/not E) The SYMPTOMS are more likely to occur when the
EVENT has occurred than when it has not.
P(E/S) < 1.00 There are instances in which the SYMPTOMS
occur when their EVENT has not occurred,
Contributory therefore the EVENT is not a necessary cause
P(S/E) < 1.00 There are instances in which the EVENT occurs

but the SYMPTOMS do not then also occur,
therefore the EVENT is not a sufficient cause

 

This is a qualitative This is a causal EVENT that occurs relatively
Proximal category, independent near the end of an etiological chain of causal

of the probability EVENTS, and which therefore occurs proximal
dependent categories to the occurrence of the SYMPTOMS

 

This is a qualitative This is a causal EVENT that occurs relatively
Distal category. independent near the beginning of an etiological chain of
of the probability causal EVENTS. and which therefore occurs
dependent categories distal to the occurrence of the SYMPTOMS

 

 

 

 

 

Figure 4
WW

. P - Probability. E - the preceding occurrence or presence of a causal EVENT.
S - the subsequent set of SYMPTOMS. and P(X/Y) - the probability of X occurring or being
present given the occurrence or presence of Y.

-.. -l1I. - II ..x::~:::.::r.t at:
.2:.::€2==13:1... .s is).
:- ...2?.T .3: Strata:

:3.....::.:«. .: {rift}
:7. «5.1.5.3.:55: 3::

A... ........ :.....

> I
.S f... ..........::.L A‘ s i I I I ......:.....:.:_.... :2:
$211.21...
. .3
e::::......:.:.. :32 t... t. 32...: c 5.... > b....:..:.::.. ...:..:.:.::.t
1- . a a.» .13.,9:2...:-:.:u-.

 

 

14

 

. r. a 28.: . .. a... a 84824. 54.5 .meaafidfi

m 2:9“.

 

     

   

«86:333.... 2 2.8
.858 cannon .550 .. n i :

  

4444a. 444.4 .

   

 

 

 

888289. .0 $33 Ens—BEES .o 520

        

   
   

   
     

 

 

.2854... 2..
Eco—4.1.03.3 3.952. 29.5.... 83 m. saw
so a . a? mo
_. a... 2. 5.: 38.32:...
ScomOmmoaoo

                  

               

        

Egon _~=o__oEc E...

92.58 ._m:o=e>=oE re... 53.6 e5 0. 8:3.er

           

           

commeaop nwocmmflono: 3:23.035 an... .o E25683 35 new Amman». 9:.
_o mESBEm _o .8 a. Eo>o 0.: o>=mao= 2.. 8. E95
32.33.30: 5:33... .303 63.45 o___ e>=muoz

   
                 

            

            
     

 

              

     
       

- $3.85....
53.85% cgmmoaou =80m.2..._ mmoam 23:30“? madam. £35m? 2:.
.o .0335 39.8230; .0 amino 2: 3 ob. o5 moi. 3268.39. .0
385555: Eu 0. Sn E E. 5...... 32.8232. .0 ounce memos". to a :8
_ ._ _ . d beset-:8 .2...on < _ at.
an... 52.85....

 

 

 
   

 

  
  

Ace—who not 2:38.352. oi
u

cemmoaou .o 39:96 20:29.: .823 5...?
.o mahmhwmwsov .. u n 1 32323045: figmwwnoflwwmgofi
mmmcmmo.waog.=oz c... .0 women. maotm> 6.55.2.8. .mzwcofimmoo

  

 

 

 

.. 1a a. mono Eco—Kai

 

 

 

 

 

mag" 21
as gene:
etc. a‘so

H‘

l'LE.Th

t0 the f3.-
intemaz a
BSteem is

Tv
are main
INLE anc
DVOQQSES
fDrmafion

15

recognizes, or at least does not exclude the possibility that "other factors such
as genetic vulnerability, norepinephrine depletion, loss of interest in reinforcers,
etc. also may be sufficient to cause depression" (Abramson et al., 1988, p. 8).

HTD proposes an etiological chain of contributory causes which
culminate in hopelessness, and therefore depression.5 The contributory cause
in this chain that is most proximal to the experience of hopelessness is the
WW regarding the cause of an
INLE. This attribution is not necessarily internal. HTD asserts that a global and
stable attribution is able to account for broad cognitive, motivational and
emotional deficits that are chronic in nature. If, in addition to being global and
stable, the attribution is also internal, these deficits will be accompanied by a
lowered self-esteem.‘5 As a contributory cause, depressogenic attributions are
not considered sufficient to cause depression, but are suggested to contribute
to the formation of hopelessness. Whether depressogenic attributions are
internal as well as global and stable is expected to impact only on whether self-
esteem is lowered.

Two contributory causes in the etiological chain proposed by HT D that
are relatively distal to the formation of hopelessness are the experience of an
INLE and the condition of one having a depressogenic attributional style. HTD
proposes that the contribution that each of these causes makes toward the
formation of hopelessness is firstly conditional on their co-occurrence, secondly

moderated by situational cues that provide information about the causes of the

. V
\.4..
‘iazagg!

lNLE. :

lNLE.

$33"
CEC'ESS
hang-gs

5'?!va

Esmp3

the 0332-.
rate... as:
become a
Woman:

The

”Edwms

16
INLE, and thirdly mediated by the actual attribution made for the cause of the

INLE.

HTD uses the concept of a cognitive Wis mechanism to
explain the necessary co-occurrence of an INLE and one having a
depressogenic attributional style for either to contribute to the formation of
hopelessness. Alloy, Clements and Kolden (1985, p. 387) state that "a diathesis
is simply a predisposition to a disorder; thus, a cognitive diathesis refers to risk
factors for disorder that are belief-based or attitudinal in nature." They continue
by stating that “stressful life events [i.e., INLE’s].... in their passive role... provide
the occasion for the operation of the cognitive diathesis... [and] in their active
role... activate or prime‘ depressogenic self or causal schemata so that they
become accessible in memory and can bias the processing of situational
information (Riskind & Rholes, 1984).“ (p. 389)

Therefore, consistent with the notion of a cognitive diathesis-stress
mechanism, HTD proposes that the role of a depressogenic attributional style
(the diathesis) in biasing actual causal attributions is not elicited without the
occurrence of an INLE (the stress), about which such depressogenic
attributions could then be made. Correspondingly, HTD proposes that the
experience of an INLE in the absence of a depressogenic attributional style
would be no more likely to elicit a biased depressogenic attribution than would
any other event.

Therefore, the functional role of the “triggering“ of a depressogenic

attributional style by an INLE in the etiological chain proposed by HTD is the

17

active biasing of the attributional process toward actually making a
depressogenic attribution for the cause of the INLE. However, HTD allows for
the fact that relevant and evident cues in the environment can moderate this
process and in some instances override the biasing effect of a depressogenic
attributional style, especially when these situational cues provide overwhelming
evidence that the INLE was due to an external, specific and unstable cause. 80,
for example, the “triggering" of a depressogenic attributional style by an INLE is
most likely to result in one actually making a depressogenic attribution (internal,
as well as global and stable) when the situational cues around the occurrence
of the INLE are characterized by a low degree of consensus (supporting the
intemality of the attribution), a low degree of distinctiveness (supporting
globality) and a high degree of consistency (supporting stability).

Ultimately then, the co-occurrence of having a depressogenic
attributional style and experiencing an INLE contributes to the formation of
hOpelessness to the degree that their co-occurrence effectively elicits an actual
depressogenic attribution regarding the cause of the INLE. Without the
formation of such an attribution, their co-occurrence would not be expected to
contribute to the formation of hopelessness, and therefore, hopelessness
depression. This is the manner in which HTD proposes that depressogenic
attributions mediate the contribution of the proposed cognitive diathesis-stress

mechanism to the formation of hopelessness depression.

 

 

(3

'¥

Cues

18
Twr nArritththHlnTh fDrinHD

Since HTD is primarily a clarification of LHTD-R, many issues relevant to
conducting an appropriate test of HTD are raised by considering the
weaknesses of past investigations of LHTD-Ft.7 Abramson et al. (1986) assert
that the empirical basis of typical investigations of LHTD-R rests upon
establishing that two conditions exist in a given sample:

1) A high proportion of depressed (or future depressed) subjects in the

sample must exhibit the hypothesized depressogenic attributional style,

and 2) A high proportion of nondepressed (or future nondepressed)
subjects in the sample must got exhibit the hypothesized depressogenic

attributional style. (p. 40)

However, HTD allows for all four possible combinations between whether
one is depressed and whether one has a depressogenic attributional style.
Figure 6 presents how HTD can account for each of these four combinations.
Figure 6 also demonstrates how other factors can figure into explaining each
possible combination, such as whether one has experienced an INLE, whether
the causal attributions one makes for the INLE are moderated by situational
cues, and whether one is experiencing a non-hopelessness type of depression.
Examination of Figure 6 clearly demonstrates that investigating the existence of
any simple relationship between depressive symptoms and one having a
depressogenic attributional style will provide insufficient information upon which
to base either a critique of, or a supportive argument for, LHTD-R (Abramson et

al., 1986; Abramson et al., 1988; Alon et al., 1988).

19

 

Presence of Presence of Experience Possible explanation of each combination which
depressive a DAS of an INLE is consistent with the characteristics of the
symptoms (diathesis) (stress) etiological chain proposed by HTD

 

l. Hopelessness depression alone

2. Hopelessness depression while also
experiencing non-hopelessness depression

YES YES YES 3. Non-hopelessness depression without also

experiencing hopelessness depression (due to

situational cues overriding the bias to form

[CELL—I?l—— a depressogenic attribution for the INLE

 

 

I h. Non-hopelessness depression and an
YES NO absence of hopelessness depression due to
incomplete diathesis-stress co-occurrence

 

 

5. Non-hopelessness depression and an
NO YES absence of hopelessness depression due to

YES
YES
inco lete diathesis-stress co-occurrence
—l:m~. .
6. Non-hopelessness depression and an
YES NO NO absence of hopelessness depression due to
incomplete diathesis-stress co-occurrence

 

 

 

7. Absence of non-hopelessness depression and
an absence of hopelessness depression due to
NO YES YES situational cues overriding the bias to form
a depressogenic attribution for the INLE
lcram. ‘1

 

 

--—J» 8. Absence of non-hopelessness depression and
NO YES NO an absence of hopelessness depression due to
incomplete diathesis-stress co-occurrence

 

9. Absence of non-hopelessness depression and

 

 

NO NO YES an absence of hopelessness depression due to
--;1 incomplete diathesis-stress co-occurrence
—-J 10. Absence of non-hopelessness depression and

NO NO NO an absence of hopelessness depression due to

 

incomplete diathesis-stress co-occurrence

 

 

 

 

 

Figure 6

 

Nggg. The reader may find it helpful to refer to Figure 1 (see page a) which provides a
schematic representation of the characteristics of the etiological chain proposed by RID.
DAS - depressogenic attributional style. HTD - Hopelessness Theory of Depression.

INLE - important negative life event. and LHTD-R - Reformulated Learned Helplessness
Theory of Depression. Regarding each 2 X 2 cell. CELL l - Depressed subjects display a DAS
which typical research on LHTD-R viewed as supporting LHTD-R; CELL 2 - Depressed subjects
do not display a DAS which typical research on LHTD-R viewed as disputing LHTD-R;

CELL 3 - Non-depressed subjects display a DAS which typical research on LHTD-R viewed as
disputing LHTD-R; and CELL A - Non-depressed subjects do not display a DAS which typical
research on LHTD-R viewed as supporting LHTD-R.

‘It is erroneous to support lflTD-R based on attributing the incidence of hopelessness
depression to the co-occurrence of depressive symptoms and a DAS (i.e., l and 2) when HTD
can account for their co-occurrence without hopelessness depression (i.e., 3 and h).
”Since HTD can explain depressive symptoms occurring in persons without a DAS (i.e., 5 and
6), it is erroneous to dispute LHTD-R based on such a finding.

cSince NTD can explain the absence of depressive symptoms in persons with a DAS (i.e., 7
and 8), it is erroneous to dispute lHTD-R based on such a finding.

‘It is erroneous to support LHTD-R based on the co-absence of depressive symptoms and a
DAS (i.e., 9 and 10) because HTD can explain the absence of each in the presence of the

other (i.e., 5. 6, 7 and 8).

 

ltd

20
In an effort to partition HT D into parts that are more readily testable 0.6.,

logically refutable), two separate components of HTD have been delineated
based on the logical characteristics of the etiological chain proposed by HTD
(Abramson et al., 1986; Abramson et al., 1988; Alloy et al., 1988; Metalsky et
al., 1982; Metalsky et al., 1987). These components of HTD, namely HTD’s
Diathesis-Stress component and HTD’s Causal Mediation component, will now

be individually addressed.

: |-.ith i - tr om- o.nnt of th H . :l: n Th . of D or: ion
The Diathesis-Stress component of HTD is intended to address the
interactive nature of the two distal contributory causes of hopelessness
depression (i.e., having a depressogenic attributional style and the occurrence
of an INLE; see Figure 5, page 14). Therefore, the Diathesis-Stress component
of HTD is the proposition that a depressogenic attributional style (the diathesis)
is essentially a latent predisposition to form depressogenic attributions (which
may contribute to the development of hopelessness depression) and that the
perceived occurrence of an INLE (the stress) can activate that predisposition.
Inherent in the Diathesis-Stress component of HTD is the proposition that
neither having a depressogenic attributional style without experiencing an INLE
(see Figure 6, page 19, cases 4 and 8) nor experiencing an INLE without
having a depressogenic attributional style (see Figure 6, cases 5 and 9) will bias

one toward making a depressogenic attribution.

C
51319 {F87
compOf‘re
hy. thee;
lNLE] p'e;

Hcv
Reactor:
censftute t
at these 5-,
0»- 49) ln Cs
Should be c
of voluntary ,
resoorfses c:
fees hopeies
13)

21
Consistent with this conceptualization of HTD, Abramson et al. (1986)

state that the first of two steps toward an adequate test of the Diathesis-Stress
component of HTD is "a demonstration that the ioteLaetjeo between the
hypothesized depressogenic attributional style and negative life events [i.e.,
INLE] predicts future depression, specifically hopelessness depression." (p. 49)
However, they continue that the second step is "a demonstration that this
interaction predicts the oomolate oonetellation of symptoms hypothesized to
constitute the hopelessness subtype of depression as opposed to only a subset
of these symptoms or symptoms that constitute other subtypes of depression."
(p. 49) In particular, Abramson et al. (1986) state that hopelessness depression
"should be characterized by at least three major symptoms: 1) retarded initiation
of voluntary responses (motivational symptom); 2) difficulty in seeing that one’s
responses control outcomes related or similar to the outcome about which one
feels hopeless (cognitive symptom); and 3) sad affect (emotional symptom)" (p.

13)

‘01] = in ma non it : a u:- =: : ofu: Drvihi- r-
l n f r
The author presently suggests that requiring the experience of an INLE,
and the condition of having a depressogenic attributional style, to interact in

predicting a W hypothesized to characterize

hopelessness depression (as required by Abramson et al., 1986) rather than to

interact in predicting W does not logically

 

 

actress ‘59 '
HTD To re:
se‘. of hose
gyms-res 2’
and t“. 12*.
However, I

Orann._hge:
s -_ I :ufl»

22

address the diathesis-stress question of the Diathesis-Stress component .of
HTD. To require that any cause of hopelessness depression predict a specific
set of “hopelessness depression symptoms“ assumes that such a set of
symptoms has been reliably associated with hopelessness depression per se,
and that this set of symptoms serves to define hopelessness depression.
However, Abramson et al. (1986) specifically state that HTD ”represents a
theomflaeod approach to the classification of a subset of depressive disorders
(see also Seligman, 1978) that is process-oriented rather than symptom
oriented.“ (p. 11). Therefore HTD emphasizes that the defining characteristic of
hopelessness depression is the orooase by which it is caused 03., the
etiological chain proposed by HTD), and not the subset of depressive
symptoms that are predicted to be displayed as a result of that process.
Therefore, it would seem more appropriate to first investigate whether the
proposed process produces depressive symptoms, and then determine whether
there is a reliable pattern in the set of symptoms produced by the process that
is distinguishable as a sub-set of a larger set of general depressive symptoms.8
Therefore, the author suggests that the first test of the Diathesis-Stress
component of HTD recommended by Abramson et al. (1986) 03., testing
whether having a depressogenic attributional style and experiencing an INLE
interact to predict depression) is both appropriate and adequate for determining
if these two factors comprise a contributory diathesis-stress component in the
etiological chain proposed by HT D (re, adequately tests the Diathesis-Stress

component of HTD). Certainly, such a test would not have direct implications for

 

mechanism
and hose- 6
CONDOhem
Di 61113163 C
an dequa‘.

testng the

23
other parts of the etiological chain proposed by HTD, but it would help resolve

whether having a depressogenic attributional style and experiencing an INLE act
independently, interactively, or act at all in contributing to the onset of

depression.

In ; s. l 1 :ol; ion oohmnnt of th Hop :|: n :o of I :pr: ion
The Causal Mediation component of HTD is intended to address the
sequential and probabilistic nature of the causes in the etiological chain
proposed by HTD (see Figure 5, page 14). in this regard, HTD segments the
etiological chain proposed by HTD into four ”links” 0.9., the diathesis-stress
mechanism, actual attributions for an INLE, the development of hopelessness,
and hopelessness depression) and a confirmation of the Causal Mediation
component of HTD would entail showing that the “links" connect in the
predicted order and fashion. More specifically, Abramson et al. (1986) state that
an adequate test of the Causal Mediation component of HTD would involve
testing the following linkages:
1) Individuals who exhibit the hypothesized depressogenic attributional
style should be more likely than individuals who do not to attribute a
particular negative life event (stress) to an internal, stable, and global
cause and view this event as important... 2) A stable, global attribution
for a particular life event and viewing that life event as important should

increase the likelihood of becoming hopeless... 3) The occurrence of

24

hopelessness should increase the likelihood of the development of

depression (specifically, hopelessness depression). (pp. 4950)

Furthermore, Abramson et al. (1986) stress that these linkages should be
shown to be probabilistically consistent with their predicted role in the etiological
chain proposed by HTD (i.e., show that the diathesis-stress mechanism acts as
a .QQfltLimeLy cause in the chain, etc.; see Figure 5, page 14, and Figure 4,

Page 13).

Inyeetigation of the Diathesis-Strees and Causal Mediation
mntfthHlnTh fDrinHTD

A considerable amount of research on LHTD-R was conducted before
the conceptually more clear HTD was available to guide the methods of such
research (for reviews, see Barnett and Gotlib, 1988; Brewin, 1985; Coyne and
Gotlib, 1983; Peterson and Seligman, 1984; and Sweeny, Anderson and Bailey,
1986). The present discussion, however, will address two studies whose
methods are more consistent with the logic of HT D.

The study of Metalsky et al. (1982) was the first to address the Diathesis-
Stress component of HTD although it did not address the Causal Mediation
component of HTD.° To test the Diathesis-Stress component of HTD, Metalsky
et al. (1982) assessed the onset of depressive symptoms in undergraduates
with versus without a depressogenic attributional style Ontemal, as well as
global and stable) after they had received an exam grade they were pleased

with (a positive life event) or an exam grade they were displeased with (INLE).

1 I1"
-

he

25
They found that internal and global (but not stable) attributional styles for lNLE’s

were correlated with depressive mood responses after experiencing an INLE
but not with depressive mood responses after experiencing a positive life event.
Williams (1985) noted that the Metalsky et al. (1982) study only examined the
relationship between having a depressogenic attributional style (internal, as well
as global and stable) and changes in depressive symptomology within positive
life event and INLE groups, rather than whether the INLE group was more likely
to show depressive symptoms than the positive life event group. Williams (1985
carried out this missing between groups comparison using data from the
Metalsky et al. (1982) study and found that the between groups difference was
not statistically significant. Williams (1985, p. 1574) stated "this implies that
mood disturbance is no more related to attributional vulnerability when students
have suffered the stress of exam disappointment than when they have suffered
no such stress.

In a later study, Metalsky et al. (1987) investigated both the Diathesis-
Stress and Causal Mediation components of HTD. In their test of the Diathesis-
Stress component of HTD, Metalsky et al. (1987) heeded the comments of
Vlfilliams (1985) and stated the following:

In line with this component of the theory [i.e., the Diathesis-Stress

component of HTD], we predicted that the content of college students’

attributional styles [i.e., depressogenic attributional style vs. depression
resistant attributional style, as measured at baseline]... would interact

significamly with the outcomes students received on a class midterm

26

exam to predict their subsequent depressive mood responses. We

further predicted that the hypothesized attributional diathesis [i.e., having

a depressogenic attributional style vs. depression resistant attributional

style, as measured at baseline]... would be significantly correlated with

students’ subsequent depressive mood responses in the presence, but
not in the absence, of receipt of a negative outcome on the midterm
exam (i.e., that the form of the interaction would be consistent with that

predicted by the theory). (p. 387)

Furthermore, Metalsky et al. (1987) conducted a preliminary investigation
of whether people might have different attributional tendencies for different types
of events and therefore show a apocfloyulnorabflty for depressive mood
responses to those types of events about which they have a depressogenic
attributional style. To study this, Metalsky et al. (1987) assessed students”
attributional styles with a revised version of the attributional style questionnaire
(ASQ) that had subscales for negative W outcomes (unrelated to
exam outcome) and negative aohiavament outcomes (related to exam
outcomes).‘° They predicted that having a depressogenic attributional style for
negative achievement events would predict students’ depressive mood
responses to a negative exam outcome while having a depressogenic
attributional style for negative interpersonal events would not.

Regarding the Diathesis-Stress component of HTD, Metalsky et al. (1987)
found that having a depressogenic attributional style for negative achievement

events predicted students’ eoduriod depressive mood responses to a negative

 

 

m

(0

tr

27

exam outcome (measured two days after the students’ receipt of their exam
grades) but not theirjmmediala depressive mood responses (measured upon
students’ receipt of their grades), and that having a depressogenic attributional
style for negative interpersonal events did not predict students’ immediate or
enduring depressive mood reactions to their grades. In their consideration of
why having a depressogenic attributional style for negative achievement events
only predicted students’ enddring mood responses to their grades, Metalsky et
al. (1987) noted:
Recent work by Weiner (1986) [aio] suggests one possible explanation
for this finding. Weiner proposed that once people perceive that an event
has occurred, they initially may experience a primitive emotional
response. Those primitive emotions, which include oaooy for success
and aad or flosoatod for failure, are labeled "outcome-dependent,
attribution-independen " by Weiner because in his view they are
determined by the attainment or nonattainment of a desired goal and not
by the causal attribution for the outcome. Weiner further argued that
following the immediate emotional reaction, a causal attribution will be
sought (particularly if the outcome is negative, unexpected, or important;
see Pittman & Pittman, 1980; Pyszczynski & Greenberg, 1981; Wong &
Weiner, 1981) and a more differentiated set of emotions then will be
generated by the chosen attribution (“attribution-dependent“).... [Hence],
once causal attributions for the negative life event have been made and

the immediate depressive mood response has begun to subside, the

 

T‘

he “Sf

a": 28’.

28
hypothesized attributional diathesis then may begin to operate (through

particular causal attributions) to predict how enduring the depressive

mood response will be.“ (p. 392)

Therefore, the findings of Metalsky et al. (1987) that exam outcome was
the best predictor of immediate depressive mood responses to exam grades,
and that the interaction between exam outcome and having a depressogenic
attributional style for negative achievement events was the best predictor for
enduring mood responses to grades, can be respectively viewed as compatible
with the notion of outcome-dependent attribution-independent and attribution
dependent emotional responses as suggested by Weiner (1985). Within the
same context, the idea that people may show a specific vulnerability for
depressive reactions to certain general types of events (1.9., achievement or
interpersonal events) is supported by the ability of having a depressogenic
attributional style for negative achievement events to predict enduring
depressive mood responses to a negative exam outcome (a negative
achievement event) while having a depressogenic attributional style for
interpersonal events was not a significant predictor of such responses.

In their test of the Causal Mediation component of HTD, Metalsky et al.
(1987) made the following predictions:

First, failure students’ [i.e., students who received a negative outcome on

their exam] attributional styles as measured [one to three weeks prior to

the receipt of exam grades] should be predictive of the particular

attributions they subsequently made for their low midterm grades.

29

Second, failure students’ particular attributions for their low midterm
grades should be predictive of their subsequent depressive mood
responses. Third, failure students’ attributional styles should not have a
direct effect, beyond that of particular attributions for the receipt of a low
midterm grade, in predicting their subsequent depressive mood
responses. (p. 387)

In making these predictions, Metalsky et al. (1987) have deviated in two
ways from examining what Abramson et al. (1986) said would comprise an
adequate test of the Causal Mediation component of HTD (see page 23). First,
rather than seeking to determine if particular attributions for an INLE predict
hopelessness and then if hopelessness in turn predicts the onset of depressive
mood responses, Metalsky et al. (1987) ignored the role of hopelessness and
predicted that particular attributions will predict depressive mood responses.
Second, Metalsky et al. (1987) have tightened HTD’s conceptualization of the
mediating role of particular attributions with their third prediction for the Causal
Mediation component of HTD. The test of the Causal Mediation component of
HTD recommended by Abramson et al. (1986) suggested an additive causal
mediation role for particular attributions in the etiological chain proposed by
HTD but did not logically exclude the possibility that attributional style might
have an additional role in the causation of hopelessness that is not mediated by
the particular attributions people make. This would have allowed particular
attributions to function as a oadial mediator (James and Brett, 1984) in the

etiological chain proposed by HTD. By suggesting that students’ attributional

 

rtsmm

The

30

styles have no power in predicting their depressive mood responses to their
midterm grades that cannot be accounted for by the particular attributions they
make for getting those grades, Metalsky et al. (1987) are suggesting that
particular attributions function as a oomoleto mediator (James and Brett, 1984)
in this process.

The study by Metalsky et al. (1987) supported all three of their
hypotheses for the Causal Mediation component of HTD, but as with their test
of the Diathesis-Stress component of HTD, the support was relative to the
prediction of students’ eoduddo depressive mood responses to negative exam
outcomes.

The logic of typical investigations of LHTD-Fl was criticized for suggesting
that LHTD-R could be supported or disconfirmed based on the correlational
relationship between the presence of depressive symptoms and the condition of
having a depressogenic attributional style in subjects in a given sample (see
page 18 and Figure 6, page 19). The issues raised in Figure 6 show that it is
necessary for research to consider factors in addition to attributional style and
depressive symptomology to draw defendable conclusions about the validity of
HTD. Alloy et al. (1988) state the following:

Three factors will influence the magnitude of differences in attributional

styles between depressed vs. non-depressed (of future depressed vs.

future non-depressed) subjects in a given sample: (1) the base-rate of

the hypothesized depressogenic attributional style; (2) the base-rate of

31

negative life events [i.e., INLE’s]; and (3) the base-rate of subtypes of

depression other than hopelessness depression.

Further examination of Figure 6 can illustrate the value of information
about these three base-rates, or baseline conditions, for research testing HTD.
Notice that explanation 1 in Figure 6 is the only instance where the observed
information can be explained by the occurrence of hopelessness depression.
Notice also that adding information about the occurrence of an INLE to
information about depressive symptoms and attributional style still does not
isolate explanation 1 from explanations 2 and 3. Only by adding further
information about the occurrence of non-hopelessness subtypes of depression
can explanation 1 be logically isolated and ultimately tested. Further, lacking
information about anyone of the three baseline conditions mentioned above, it
is impossible to logically isolate hopelessness depression and test HTD
because any relationship between the other two baseline conditions and
depressive symptomology could be explained within the logic of HTD (see
Figure 6, page 19).12

While Metalsky et al. (1987) took considerable care to attempt a fair and
appropriate test of the Diathesis-Stress and Causal Mediation components of
HTD, they did not measure the W of students’ experiences of
INLE’s (e.g., other lNLE’s they may have experienced prior to their participation
in the study) or address the issue of non-hopelessness subtypes of depression

in their study. Because of these gaps in the logic and design of their study,

Metalsky et al. (1978) can not (and do not) claim to have folly and logically

32

isolated a hopelessness depression response in students, or to have similarly
tested HTD. For example, with regard to not addressing the baseline condition
of non-hopelessness depression in their sample, Metalsky et al. (1987, p. 393)
state that "similar to other investigators, we treated depression as a unitary
phenomenon and did not search for hopelessness depression." Therefore, they
recognized that their results were interpretable only with reference to depression
in general and did not lend eoeofic support to the notion of nopoIoasoasa
depression.

The potential confounding impact of Metalsky et al. (1987) not measuring
the baseline condition of students’ experiences of lNLE’s on interpreting their
results is less clear. Since Metalsky et al. (1987) measured temporally
constrained onemoee in depressive mood, it is not likely that any students’
experiences of INLE’s prior to the experiment exaggerated the increases in
depressive mood that were observed. However, it is possible that the observed
increases in depressive mood could have been curtailed to some degree by
students mood states as they related to earlier experiences of lNLE’s (e.g., they
were already depressed and did not become much more depressed after the
receipt of a negative exam grade) or that their mood reactions were otherwise
moderated by earlier experiences of lNLE’s. In any instance though, future
studies of HTD can avoid the concern of how the experience by subjects of
other lNLE’s prior to an experiment might confound such research by directly

addressing the issue.

33
r'i;l ‘IIr-I I Ifth Rufrm MN _:.m:- H-III- n Thu If

I.” a4 'on HTI-R =n- h Haul: n :o of error: ion HTI

Since HTD is in small part a modification of LHTD-R, and in large part, a
clarification of LHTD-R, criticisms of both theories are relevant to this study.

In an early review of investigations of the role of cognition in depression,
Coyne and Gotlib (1983) concluded that “neither Beck’s nor the learned
helplessness model of depression has a strong empirical base." (p. 472) In
particular, Coyne and Gotlib (1983) point out: (a) the failure to obtain significant
depressed-non-depressed group differences for attributions for success and
failure, (b) depresses-non-depressed group differences of disappointing
magnitude for attributions for hypothetical events, and (c) inconsistent
depressed-non-depressed group differences for attributions for stressful life
events. They also point to failures to predict subsequent depression from
current cognitions. They draw attention to the study of Lewinsohn, Steinmetz,
Larson and Franklin (1981) which found that:

People who are vulnerable to depression are not characterized by stable

patterns of negative thinking of the type postulated by the cognitive

theorists. Apparently people change their expectancies and subscribe to
irrational beliefs as a result of being depressed, and these cognitive

changes reverse themselves as the individual recovers. (p. 218)

 

.Ifi DI~
Hes-

It! '1
W“,
n It-
P'a- QC“

00358

34

Considering the results of such studies of cognitive vulnerabilities to depression,
Coyne and Gotlib (1983, p. 499) state:

It is likely to be difficult to distinguish empirically a hypothesis concerning

latent cognitive factors from the traditional assumption that when

confronted with certain internal and external stimuli, some people
become depressed, and negative thinking is simply one characteristic of
this state.

Coyne and Gotlib (1983, p. 501) thus state that "The field in general
would benefit from the recognition and articulation of rival hypotheses to explain
why depressed persons make negative self-reports." They suggest that
particular issues worth investigating are the environmental antecedents and
consequences of depressed persons’ negative verbalizations and the coping
strategies and behaviors of such individuals.

In an ensuing dialogue stimulated by the Coyne and Gotlib (1983) article,
Segal and Shaw (1986a, 1986b; representing Beck’s Cognitive Theory of
Depression) and Coyne and Gotlib (1986) debate the above issues and also
discuss the implications of conceptual modifications and changes in emphases
in cognitive theories of depression. The result of the dialogue is summarized
well by Segal and Shaw (1986b, p. 707) when they state "While many of the
criticisms raised by Coyne and Gotlib are valid and shared concerns, the
divergence in views is most apparent when solutions for these difficulties are
considered.” While all parties seem to advocate a diversification of focus in

future research, Coyne and Gotlib (1986, p. 695) maintain that "the central

35

defect in current cognitive approaches to depression is their inattention to the
difficulties faced by depressed persons in their everyday environments, how
they cope, and with what consequences." Segal and Shaw (1986b) advocate
the emphasis of the diathesis-stress nature of cognitive models of depression
over investigations of the cognitive diathesis alone, and also consider the future
investigation of cognitive constructs that are more interpersonal in nature.

In a review of studies of the relationship between attributions and
depression, Brewin (1985) considered five causal models. In the symptom
model, a bad event leads to the onset of depression which leads to depressive
attribution. In the Model, a bad event leads to depressive attributions
which leads to the onset of depression. In the mlneranilm model (a diathesis-
stress model similar to the etiological chain proposed by HTD), a bad event
interacts with a depressive attributional style to produce depressive attributions
which leads to the onset of depression. In the W. a bad event
leads to the onset of depression and depressive attributions or a depressive
attributional style can lead to the maintenance of depression which would
otherwise remit. In the W, a depressive attributional style leads to
the onset or maintenance of depression. Brewin (1985) concluded that the
research to date provided support for the symptom, recovery, and coping
models, but did not support the two models most similar to LHTD-R, namely the
onset and vulnerability models.

The criticisms of LHTD-R just presented are valid and have resulted, in

part, in the more clear restatement and reformulation of LHTD-Fl as HTD. As

 

~
“EVE a-

‘v'w‘ by
I

”C

re

36

discussed earlier (see pages 18-21 and Figure 6), typical investigations of
LHTD-R base their findings on simple depressed-non-depressed group
differences (like those mentioned by Coyne & Gotlib, 1983) and thus have
ignored the diathesis-stress nature of LHTD-R. Such studies drew attention to
the need to state LHTD-R more clearly. Also, consistent with the findings of
Brewin (1985), and as discussed earlier on pages 24-32, investigations that
have addressed the diathesis-stress nature of LHTD-Fl (or of HTD) have still
failed to address all of the methodological implications of the theory.

Barnett and Gotlib (1988) offer an excellent analysis of the
methodological issues that pertain to the identification of psychosocial factors
that may oadse depression, and of how many studies do not adequately
distinguish the antecedents, concomitants and consequences of depression.
This concern of Barnett and Gotlib (1988, p. 97) is clearly stated:

The failure in most studies to evaluate the interaction between initial

symptoms and the predictor variable... confounds attempts to link the

predictor with the actual onset of depression. For example, a measure of
cognitions may be a significant predictor of subsequent level of
depression, but because subjects differ in their initial symptom levels, it is
not clear whether cognitions are predicting the onset, exacerbation, or
remission of depression in a group of subjects (of. Hammen, Mayol,

deMayo, & Marks, 1986).

Barnett and Gotlib (1988) advocate for the use of prospective research to

investigate etiological factors in depression. They suggest that a memoflid

 

E
’I
9;- pr:

rSitar to is

”CISGIy c

meaSUremQ

37
oase;oontLol_desion or a two-wave panel design ("in which a psychosocial
variable is used at one time to predict subjects’ subsequent levels of
depression." p. 97) can be used effectively, particularly when the distinction
between remitted depressives and normal controls can be maintained in the
non-depressed portion of a subject sample.

When considering research on the relationship between attributional style
and depression, Barnett and Gotlib (1988) point to a large body of research that
is critical of LHTD-R along the same lines as the critique presented by Coyne
and Gotlib (1983), including later studies that also failed to show that
attributional style alone predicted various aspects of the course
and symptoms of depression. They also note that “Remitted depressives did not
exhibit more attributional biases than control subjects.“ (p. 103)

However, Bartlett and Gotlib (1988, pp. 103, 106) note that:

There have been no adequate tests in adults of the full diathesis-stress

model proposed by Peterson and Seligman (1984). The positive results

obtained by Metalsky et al. (1987) in their study of negative mood are
promising, and suggest that the interaction of attributional style and
negative life events should be investigated in prospective research using
proper measures of depression to explore the issue more fully.

The Peterson and Seligman (1984) model that Barnett and Gotlib (1985)
refer to is similar to that proposed by HTD, though less conservative, and is
concisely summarized by Brewin (1985, p. 303) as requiring ”(a) prior

measurement of attributional style, (b) the occurrence of a bad event, (0) an

HTD. are
when the
35.413 of
Mappers.

[Ital

tar-Cat:

I‘l’hile the
Vulnerab;
have bee
the iernm
egress]:
addi'esse
bellow u

Third: as
1988) an:

38
attribution for the event in line with the previously determined style, and (d) the

ability of attributional style to predict depression following the event." Barnett
and Gotlib (1988) note that the use by Metalsky et al. (1987) of an adjective
check list rather than a measure of depression as their dependant measure
limits the comparability of that study to other depression research. Therefore,
they apparently believe Metalsky et al. (1987) did not measure depression and
therefore failed criterion (d) stated above.

The implications of this literature that are critical to LHTD-R, and hence
HTD, are derivable from the conclusions of Barnett and Gotlib (1988, p. 97)
when they state "The review suggests that whereas there is little evidence in
adults of a cognitive vulnerability to clinical depression, disturbances in
interpersonal functioning may be antecedents or sequelae of the disorder." The
implications are first, that future investigations of cognitive vulnerability to
depression must be designed to provide conclusive tests of such theories.
While they are correct in saying there is little supportive evidence for a cognitive
vulnerability to depression, they are also correct in not saying that such theories
have been conclusively tested and refuted. Their concerns regarding isolating
the temporal relationship of cognitions such as depressive attributions and
depression pertain to this implication for research design. Second, research that
addresses these important methodological concerns should be extended
beyond undergraduate student samples to samples of the clinically depressed.
Third, as was a conclusion of the dialogue between Coyne and Gotlib (1983,

1986) and Segal and Shaw (1986a, 1986b), other factors such as interpersonal

 

fast (ha

Barrie:

39

functioning are being conclusively shown to play an etiological role in the onset of
depression, thus warranting the development of multifaceted models of depression
that incorporate interpersonal, environmental, and if they prove to be of utility,
cognitive factors.

The implications of this literature for this study must be viewed in light of the
fact that the study was designed and conducted prior to obtaining the insights of
Barnett and Gotlib (1988). As will be seen in the Method (see pages 45-56), this
study uses a synthesis of the premorbid case-control and two-wave panel research
designs advocated by Barnett and Gotlib (1988). It also provides a method that is
consistent with the full methodological implications of HTD. However, this study 1)
was not designed to distinguish between remitted depressives and normal
controls, 2) used an adjective check list for mood as the dependent measure, and
3) used a sample of premorbid undergraduate students, some of which showed
a change in mood, rather than a sample of premorbid clinical depressives. Barnett
and Gotlib (1988) therefore, would have grounds based on these points to criticize
this study as also providing “little evidence in adults of a cognitive vulnerability to
clinical depression“ (Barnett and Gotlib, 1988; p. 97) even if all its hypotheses were
supported. However, they would likely acknowledge that this study allowed for a
strong test of whether some people may have a cognitive vulnerability to
experiencing an enduring depressive mood response (short of a clinical

depression) in response to an INLE.

 

 

1)-

4O

Au'3AIPI'I'OI-3th30 :nIIslz : l=I If|:I: ion TD

The present study was designed to be another step, beyond those taken
by Metalsky et al. (1982) and Metalsky et al. (1987), toward a wholly
appropriate and adequate test of HTD. The primary focus of this study is to
modify the methods used by Metalsky et al. (1987) in a manner that allows for
the logical identification of depressive mood responses of the hopelessness
type. In order to do so, the challenge to be met is to obtain information on the
baseline condition of students types of attributional styles, experiences of
INLE’s, and the incidence of non-hopelessness subtypes of depression. The
. measurement of the baseline condition of attributional styles and INLE’s is
described in the methods, but the operationalization of the baseline condition of
non-hopelessness subtypes of depression merits discussion here.

A serious obstacle to obtaining information on the baseline incidence of
non-hopelessness depression is that both hopelessness depression and non-
hopelessness depression require information about each other to be logically
identified. That is, to isolate hopelessness depression, one needs information on
the baseline condition of non-hopelessness depression, while non-hopelessness
depression is most generally defined as the presence of depressive
symptomology in the absence of hopelessness depression. This study
addresses the apparent circularity of these definitions by screening al_I
depressed subjects from its sample, thereby assuring that there is a zero
incidence of non-hopelessness depression (or of any subtype of depression) in

the sample.

41

So by adding a measure of lNLE’s and a measure of depression during
the baseline period, subjects who are depressed and/or have experienced a
recent INLE can be screened out of the sample, yielding a zero incidence for
. each of these factors in the sample. Given such a sample and using methods
otherwise comparable to those of Metalsky et al. (1987), a similar test of the
Diathesis-Stress and Causal Mediation components of HTD is possible with the
benefit of being able to logically identify whether subjects’ depressive mood
responses are of the hopelessness type.

For this reason, the predictions of this study are very similar to those of
Metalsky et al. (1987). However, the following are differences in the predictions
used in this study. First, all predictions are stated in the context of having
determined the baseline condition of subjects’ attributional styles, experiences
of INLE’s, and the incidence of non-hopelessness depression. Second, in
accordance with the findings of Metalsky et al. (1987), mood related predictions
are made with regard to subjects’ enduring depressive mood responses. Third,
two additional predictions are added to the test of the Causal Mediation
component of HTD that are designed to begin to explore the relationship
between feelings of hopelessness and hopelessness depression. .

‘lhe study of Metalsky et al. (1987) did not address the experience of
hopelessness as a component of the etiological chain proposed by HTD in their
test of the Causal Mediation component of HTD. While this has been noted as a
weakness in that test of the Causal Mediation component of HTD (see pages

31-32), steps toward correcting that weakness in this study were limited to

 

.‘(Iflh h "

.5 a"

arid a 59'

vyvg N

42

taking two measures of subjects’ hopelessness, one during the baseline period
and a second after all other data were collected. This is because the author has
chosen to maintain a high degree of comparability between the methods of this
study and those of Metalsky et al. (1987) so that the impact of the sample
selection methods of this study might be better isolated. Any measures of
hopelessness between these two implementations could have caused
experiential differences for subjects that would have limited the comparability of
the two studies. Therefore, the two hypotheses for the Causal Mediation
component of HTD regarding hopelessness are limited to elaborating on the
relationship between feelings of hopelessness and the experience of a
depressive mood response of the hopelessness type, and do not examine the

functional role of hopelessness in the etiological chain proposed by HTD.

'hi-r mnnH th Dithi-r H“°’
Thus, regarding the Diathesis-Stress component of HTD, this study
predicts that among a sample of students who, at baseline, are not depressed,
and report no recent INLE’s:
Diathesis-Stress H‘) There will be an interaction between students”
attributional styles as measured before their midterm exam and the outcomes
students received on their exams that will predict their subsequent enduring

depressive mood responses.

 

De
wi.’ be a S
depresso;
espresswe

.5?
mi; not be .
students he

51106715 8.".

SCOTBS.

43

Diathesis-Stress H2) For students who receive a low exam grade, there
will be a significant positive correlation between the state of students having a
depressogenic attributional style at baseline and increases in students enduring
depressive mood subsequent to the receipt of their exam scores.

Diathesis-Stress H3) For students who receive a high exam grade, there
will not be a significant positive or negative correlation between the state of
students having a depressogenic attributional style at baseline and increases in
students enduring depressive mood subsequent to the receipt of their exam

SCOfBS.

lMitin mnntH th lMitinH‘

Regarding the Causal Mediation component of HTD, and given the same
sample, this study predicts:

Causal Mediation H‘) For students who receive a low exam grade, there
will be a significant positive correlation between students’ attributional styles as
measured before their midterm exam and their respective attributions for their
performance on their midterm exam.

Causal Mediation H2) For students who receive a low exam grade, there
will be a significant positive correlation between students’ respective attributions
for their performance on their midterm exam and students’ enduring depressive

mood responses.

 

vhibe
fiytfe

need I

bۤ~95
556ml

Endup;

44

Causal Mediation H”) For students who receive a low exam grade, there
will be a significant positive correlation between students’ respective attributions
for their performance on their midterm exam and students’ enduring depressive
mood responses that is independent of any correlation between students
attributional styles as measured before their midterm exam and their enduring
depressive mood responses.

Causal Mediation H‘) There will be a significant positive correlation
between students’ feelings of hopelessness (measures two days after the
receipt of their exam grades) and students’ enduring depressive mood
responses.

Causal Mediation H5) There will be a significant positive correlation
between students’ increases in feelings of hopelessness from baseline (Time 1,
see method below) to two days after the receipt of exam grades and students’

enduring depressive mood responses.

METHOD

meets

The subjects for this study were 43 undergraduate students at Michigan
State University, (23 enrolled in an introductory psychology course and 20
enrolled in a psychology statistics course). The initial subject pool consisted of
343 who took the first midterm exam in one or the other class. Attrition due to
voluntary nonparticipation in the study or incomplete participation resulting in
missing data left 124 subjects. The selection of students out of the sample for
having a high depressive mood or for having experienced an INLE (see pages
4748) yielded the research sample of 43 subjects.‘3

It was assumed that students’ enrolling in these classes would have little
or no knowledge of the measures or theories used in this study. Therefore, this
sample was assumed to be free of any bias that might result from subjects
having such knowledge. Students’ participation in this study was voluntary,
although it did fulfill a requirement to gain formal exposure to the research
process for students in the introductory psychology course, and it earned
bonus points toward their final grade for students in the psychology statistics

class.

45

 

Adject‘ve
376 Beck
Ema-.94,
80' 'meas
iiiStantanec
M81339 e1
MAACLRel

Wad to thee

46

Measures

The degree to which students had a depressogenic attributional style
toward making global and stable attributions for negative achievement events
was measured with the revised version of the Attributional Style Questionnaire
(ASQ, see Appendix A) used by Metalsky et al. (1987). Similarly, the nature of
the actual attributions students made for their performance on their exam was
measured using the Particular Attribution Questionnaire (PAQ, see Appendix B)
developed by Metalsky et al. (1987). The PAQ also asks students whether they
were ”happy" or "not happy" with their exam grade and was used to measure
students’ subjective evaluation of their exam outcome.“ Metalsky et al. (1987)
reported that the A80 had a reliability coefficient of .79 for measuring the
degree to which students had a depressogenic attributional style, and that the
FAQ had a reliability coefficient of .81 for measuring the degree to which
students’ respective attributions for their exam outcomes were depressogenic.

Metalsky et al. (1982) recommended the use of the Multiple Affect
Adjective Check List - Today Form (MAACL; Zuckerman & Lubin, 1965) over
the Beck Depression Inventory (BDI; Beck, Ward, Mendelson, Mock, and
Erbaugh, 1961) for measuring subjects’ t_rens_ient_ levels of mood because the
BDI “measures more enduring symptoms of depression rather than
instantaneous levels of depressive affect“ (Metalsky et al., 1982, p. 614).
Metalsky et al. (1987) also used the MAACL. The depression scale of the
MAACL-Revised (MAACL-R, see Appendix D; Zuckerman & Lubin, 1985) was

~ used to measure transient mood in this study because it retained the temporal

E.
e I‘Vl'ltil' '

 

I)
(It
I 1

47
sensitivity of the MAACL, but also has been restandardized using a sample

more similar to the subjects in this study 0.9., undergraduate students) and now
takes subjects’ gender and response frequency into account in determining
scores. Zuckerman and Lubin (1985) reported that the reliability (Alpha) of the
depression scale on the MAACL-R Today Form among college students ranges
from .74 (n=245) and .75 (n=60) to .80 (n=536). They also reported several
studies that support the convergent and discriminant validity of the MAACL-R
scales in college student samples.

The BDI (see Appendix E) was used to identify those subjects who
displayed general depressive symptoms at baseline (i.e., BDI score > 9). In
their recommendations for the use of the BDI, Kendall, Hollon, Beck, Hammen
and Ingram (1987) reserve the use of the term "nondepressed" for subjects who
score 0-9 on the BDI. Depressed subjects were selected out of further analyses
for reasons already discussed (see pages 40-41). Beck et al. (1961) reported
the internal split-half reliability of the BDI as .86 rising to .93 with a Spearman-
Brown correction. They also reported the correlation between BDI scores
clinicians’ ratings of depth of depression to be .65 (n=226) and .67 (n=183).

The Beck Hopelessness Scale (BHS, see Appendix F; Beck, Weissman,
Lester, and Trexler, 1974) was used to measure students’ feelings of
hopelessness. Beck et al. (1974) reported a reliability coefficient (Alpha) of .93
for the BHS. They also reported that the BHS correlated .74 with clinical ratings

of hopelessness and .62 with attempted suicide.

48
The Life Events Inventory (LEI, see Appendix G; Cochrane & Robertson,

1973) was used to identify those subjects who had experienced an important
negative life event (INLE; i.e., endorsed any event on the LEI as “extremely
negative") in the last six months. Such subjects were selected out of further
analyses for reasons already discussed (see pages 40-41). However, the LEI
was designed to measure general life stress due to cumulative life events. Since
this study used the LEI only to identify the experience of any singular extremely
negative life event, there is no directly applicable reliability data available to
report. However, the LEI appeared to offer a valid sampling of potentially
stressful events including several specifically related to experiences common to
students. The LEI also offers opportunities to report stressful events not offered

elsewhere on the inventory.

Emcedute

Although there are several differences between this study and that of
Metalsky et al. (1987), an attempt was made to replicate the procedure
employed by that study where possible. The following procedure reflects as
much of a procedural replication as was possible based on the constraints of
subject availability, class and exam schedules and information from the
Metalsky et al. (1987) study and from personal communications with Gerald I.
Metalsky.15

All measures in class were administered in large lecture halls with

terraced seating and no windows. These settings may have created a moderate

‘ggggfih"”5

:EflGC
in ages
The var

IT‘TE'ICE.‘

49

demand characteristic to participate in the portion of the study conducted in
class both through pressure to conform to the activities of a large group and
through the lack of an alternative activity being provided to participate in while
others completed questionnaires. However, such a demand seemed likely to
produce a larger and more representative sample of the class rather than
produce a systematic sampling bias.

All measures out of class were administered in single sessions within a
period of about two weeks in one of several moderate size classrooms similar
in appearance and location or in the subjects’ lecture hall after class had ended.
The variety of times and locations available for giving these measures was

intended to increase participation by appealing to the subjects’ convenience.

[met

Students were solicited to participate in what was designed to be
perceived as two independent studies (i.e., "The Mood Study,” and the
“Personality and Style" study; see Figures 7, 8 8. 9), and completed consent
forms for each study before their participation began.16

The Mood Study consisted of all administrations of the MAACL-R. A
rationale for The Mood Study which was sufficiently vague not to suggest the
true nature of the experiment or to create a demand characteristic for
participant response styles (see Appendix H) was given to the students verbally
in class by a graduate student. To help mask the temporal relationship between

MAACL-R administrations and the exam, the MAACL-R was administered in

 

PM

L"I-‘.l'

LS

Time 1: Baseline Measures

.Administered - . . - .Administered ' . . Time 2: Time 3:
out oflclass in class . immediate enduring

 

       
      
      

   

assessment assessment 8988 ODS

 

   
    

  

 

 

session in a over 3-4 weeks depending Subjects Exam grades MAACL-R
2 week period on which class the take are returned BHS
student is in exam MAACL-R
j l
_ASQ, LEI. BDI, BHS 5-7 MAACL-Rs
Pigure 7
Ihs_9sIalsss_§tndr_aa_£gnduetsd

8251- For Figures 7, 8 and 9: Time 1 - A period beginning in the second week of classes in
the term. Time 2 - ten minutes after exam grades are returned, Time 3 - two days after
Time 2. ASQ - Attributional Style Questionnaire, BDI - Beck Depression Inventory,

INS - Deck Hopelessness Scale, LEI - Life Events Inventory, MAACL-R - Hultiple Affect
Adjective Check List-Revised (Depression Scale), PAQ - Particular Attributions
Questionnaire.

Time 1: Baseline Measures

 

.Administered ' ' . Time 2: Time 3:
in ilass . immediate enduring

res OBS.

   
 
    
  

 
 
  

assessment sessions
over 3-4 weeks depending

  

Subjects Exam grades

   

   

 

on which class the take are returned
student is in exam MAACL-R
5-7 union-as
Figure 3 .
W
Time 1: Baseline Heasures

.Administered ' ' ' ' ' ' ' ' ' ° ° ' ' . Time 2: Time 3:
out of class . immediate enduring

response

   
     
 

 

  
 
 

ISIOIIIQDC

  
 

 

session in a Subjects Exam grades BHS
2 week period take are returned
4 exam PAQ
ASQ, LEI, BDI. BHS
Figure 9

 

b‘SE‘Ore 1

51

every class session beginning the second week of the term until two class
sessions after the exam, with the exception of the day of the exam. This
resulted in five baseline MAACL-R administrations in the psychology statistics
course and seven in the introductory psychology course (the exam in the latter
occurred one week later). MAACL-R administrations occurred as follows. At the
beginning of class, an MAACL-R Today Form was distributed to each member
of the class. Subjects were asked to complete the form upon receipt of it
Gdentifying themselves only with their student ID. numbers), and when it had
been determined that all subjects had completed their form, they were then
asked to return the form to the end of their aisle to be collected.

Solicitation for participation in the Personality and Style study was done
by a different graduate student and on a different day than for The Mood Study
(see Appendix I). The Personality and Style study consisted of the remaining
non-MAACL-R measures (i.e., the ASQ, BDI, LEI, FAQ and both administrations
of the BHS). Of these, only the ASQ, BDI, LEI and the first administration of the
BHS were given in the assessment session out of class. All four measures were
administered in single sessions, and subjects were permitted to attend only one

of the many available sessions out of class.

The 2.
At Time 2 0.9., five days and the first class meeting after subjects took
their exam), exam grades were posted outside of each classroom ten minutes

before the start of class so students could view them as they entered the room.

 

(I)
I ‘¢
(I)
2!.
(I)
r+

52

At the start of class, the instructor announced that the exam grades were
posted and that students should check their grade if they had not yet done so.
After the remaining students checked their grade, the MAACL-R was
administered. Immediately following, the PAC was administered in the same

manner as the MAACL-R.

I ImQ 3-

At 'I'Ime 3 (i.e., two days and the next class meeting after Time 2), the
MAACL-R was administered. Immediately following, the second administration of

the BHS was done in the same manner as the MAACL-R.

Sunseddent to Time 3.

The true rational for this study and the rational for any deceptions was
presented in both written and verbal form to both classes (see appendix J).
Subjects were given the opportunity to ask questions of the experimenter both

in and out of class.

5! l' !' | I! !
Most of the hypotheses in this study were tested using a special variation
of a multiple regression/correlation analysis approach to the analysis of
covariance called the analysis of partial variance (Cohen & Cohen, 1983, pp
402-423). Multiple regression/correlation analysis involves a wide variety of

procedures that determine the specific relationship between the variance

SCJCG
p hrof“
.83. .5:

'

Mtf
WV

3. is
t: the E
0'6383
rare: -:
7643? 3r

‘:~fir
‘uvtv

53

associated with two or more independent variables and the variance associated
with a dependent variable. A more specific application of multiple
regression /corr9lation analysis involves the statistical control of irrelevant
sources of variation by using a setwise hierarchical multiple
regression/correlation analysis as a special Instance of the analysis of
covariance.

In their discussion of a multiple regression/correlation analysis approach
to the analysis of covariance, Cohen and Cohen (1983, pp. 379-402) provide a
procedure for partialing the variance associated with a set of independent
variables (i.e., the covariates) from that of the dependent variable so the
relationship between another set of independent variables (i.e., the research
factors) and the remaining variance of the dependent variable can be
determined. They note that for such an analysis of covariance to be valid, 9 test
of the assumption of homogeneity of regression of the dependent variable on
the covariates must be conducted because the calculations used to adjust the
dependent variable for the effects of the covariates assume that the slope of
their regression lines are equal. This serves to ensure that when the dependent
variable is adjusted for the effects of the covariates: 1) meaningful variance from
the dependent variable is not lost, and that 2) irrelevant variance from the
covariates does not remain and contaminate the adjusted dependent variable.

Thus, the covariates which are presumed to carry irrelevant variance are
entered into the multiple regression/correlation analysis equation first to partial

their distorting influence out of the equation. By definition then, the adjusted

 

COVE?
hono

3025:;

54

dependent variable correlates 0.00 with the set of covariates. Then the research
factors are entered into the equation in whatever order or manner is logical
given the experimental design so their relationship to the adjusted dependent
variable can be determined (i.e., to test the research hypotheses).

Finally, the interaction terms between the covariates and the research
factors are entered into the equation to test the assumption of homogeneity of
regression. If this last set of interaction terms are found to account for a
significant portion of the remaining variance of the dependent variable (after the
dependent variable is adjusted for the covariates and the research factors), the
assumption of homogeneity of regression is violated and the analysis of
covariance must be considered invalid. Otherwise the assumption of
homogeneity of regression is supported and the analysis of covariance is
considered valid and interpretable.

The analysis of partial variance is a special application of the multiple
regression/correlation analysis approach to the analysis of covariance.
Typically, the independent variables in an analysis of covariance (i.e., often the
covariates and more often the research factors) are represented with regard to
group membership. The distinguishing characteristic of the analysis of partial
variance is that it allows for the use of odan_t'rmiye independent variables as
covariates and research factors in an analysis of covariance. That is, in the
analysis of partial variance, the independent variables that form the covariates

and the independent variables that form the research factors may be of my

Inorvg:
g.u U“
M r 3"
4 ‘3 E’ o

V3752 C

38m:

matter
33 34.2

teat?

55

mm (i.e., their values may be represented on nominal, ordinal, interval or
ratio scales).

Cohen and Cohen (1983, p. 407) note that "measurement error may
decrease or increase or even change the sign of a denial relationship" and that
therefore the reliability of the independent variables used in an analysis of partial
variance must be evaluated. They state that “we cannot stress too strongly our
conviction that the best way to deal with this problem is to finesse it entirely by
using independent variables whose reliability is, if not perfect, at least high," (p.
411) and that "failing to correct when reliability is .8 or more will usually not
matter much, but the risk is very great when it is as low as .5 or .6." (p. 409)

So quantitative independent variables that are determined to have a high
reliability may be used in an analysis of partial variance.

A special application of the analysis of partial variance involves the study
of change, which is how the analysis of partial variance was used in this study.
To study change with the analysis of partial variance, one simply uses a pre-
intervention measure of the dependent variable as the covariate to be partialed
out in the first step of the regression equation (Cohen and Cohen, 1983). Thus,
by adjusting the dependent variable for pre-intervention scores on the same
measure, any remaining variance in the adjusted dependent variable will
correlate 0.00 with the pre-intervention measure and can be viewed as due to a
change from that condition. Consequently, any relationships determined to exist
between any research factors and the adjusted dependent variable can be

viewed as related to the _c_h_anoe that occurred between the pre and post

\e-e‘

' 1|- “I!
;R
vpvd \-

a ‘Anrv
VUIU'U

56

intervention conditions. Therefore, since this study is concerned with changes in
depressive mood, all post intervention measures of depressive mood were
adjusted for an estimate of students’ baseline level depressive mood before the
research factors (i.e., measures of depressogenic attributional style, exam
outcome, students’ actual attributions for their performance on their exams, and

hopelessness) were entered into their regression equations.

 

I)
It)
(I)

USEd ll

RESULTS

All data analyses were done using the SPSS-X Data Analysis System,
Release 3.0 (SPSS Inc., 1988), on the mainframe computer at Michigan State
University. Although the hypotheses of this study focused on the prediction of
students’ enduring (i.e. two days between the receipt of exam grades and Time
3) depressive mood responses to the receipt of their exam grades,
corresponding analyses of students’ immediate 6.9., Time 2) mood responses

will also be presented for comparison.

f mtinn frnn ' f ilvrin

The analysis of partial variance assumes that the independent variables
used in an analysis either have a high reliability or that their has been a
correction for measurement error. The reliability coefficients reported here used
data only from the 43 students used in the data analyses. The reliability
coefficient for the Attributional Style Questionnaire’s (ASQ) measurement of
students’ attributional styles toward making global and stable attributions for
negative achievement events was .86. The reliability coefficient for the Particular

Attribution Questionnaire’s (PAQ) measurement of the degree to which students

57

reset:

tore
we'e

9w

.MU

A‘s:
lit-.4

 

 

58

formed a global and stable attribution for their exam performance upon the
receipt of their exam grades was .87.17 Since ASQ scores and PAQ scores
were calculated by taking the simple average of scores on their globality and
stability subscales, these coefficients were calculated to reflect the reliability of
the linear combinations of those subscales (Nunnally, 1978).

Students’ baseline moods were estimated by taking the mean of Multiple
Affect Adjective Check Ust-Revised (MAACL-R) depression scale scores from
several baseline implementations of the MAACL-R. The reliabilities (Alphas) of
MAACL-R depression scores for each of the baseline assessments were .95
(n=22), .59 (n=23), .60 (n=41), .84 (n=40), .83 (n=34) and .86 (n=40).18

The analysis of partial variance also requires that a test for the
assumption of homogeneity of regression of the dependent variable on the
covariate independent variable be conducted for every analysis. This
assumption was supported for every analysis conducted in this study. That is, in
no instance did the interaction terms between the covariate and the research
, factors account for a significant portion (p<.05) of the remaining variance of the
dependent variable after the dependent variable was adjusted for the covariates
and the research factors.19 So as not to detract from results for the tests of
research hypotheses, the results of the tests for the homogeneity of regression
are presented in Table A, Appendix K rather than in the tables describing the
analysis of partial variance analyses. These tests would otherwise be seen as

the final step in each analysis of partial variance.

i

(I)
(I
u

59

A brief discussion of the statistics used in reporting the results of the
hypothesis tests (i.e., R2, srz, pr, and prz; Cohen & Cohen, 1983) is presented
here. This discussion is conducted with reference to Table 1 [which reports the
statistical analysis of the first Diathesis-Stress hypothesis (Diathesis-Stress H‘);
see page 64) to give an example of the application of these statistics in this
study. Table 1 reports an analysis of partial variance designed to determine if
students’ ASQ scores interacted with their exam outcomes to uniquely and
significantly predict their MAACL-R depression scores at Time 3.

The R2 statistic is the proportion of the dependent variable variance On
this case, Time 3 MAACL-R depression score) that is shared with afl of the
optimally weighted independent variables that are in a regression equation in a
single step. Therefore, in the first step of the equation, students’ baseline
MAACL-R depression scores accounted for 16% (R2=.16) of the variance in
their Time 3 MAACL-R depression scores. In step 2, the research factors of
A80 score and exam outcome were simultaneously entered into the equation
so their individual effects would be partialed out before testing the effect of their
interaction in the third step. The R2 for the second step, .18, represents the
proportion of the variance of Time 3 MAACL-R depression score that is
accounted for by ASQ score, exam outcome and baseline MAACL-R
depression score.

The sr2 statistic is the proportion of the dependent variable variance that

is accounted for by all of the independent variables entered into the equation in

2 sins".
am:
are! S

nor-rein.
navy.

Thorn?
= 32‘,
5336

R '4‘,
Up.

vvl

fed

l’9cr
pg»?

60

a single step, less the proportion of the dependent variable variance that is
accounted for by all of the independent variables entered into the equation in all
m steps. Therefore, sr2 represents the neti_nc_r_ease in R2 due to the addition
into the equation of the independent variables just entered in a given step.
Therefore, given an R2 for step two of .18, the er2 of step two is .02 (.18 -.16
:02). The sr2 for the first step of an equation is always equal to the R2 for the
same step since 0.00% of the dependent variable variance is accounted for
before the first step. An F value is determined relative to the sr2 for a given step
as an indication of the significance of the net increase in R2 due to the inclusion
of the independent variables in that step.

The pr statistic is the eorrelation between an independent variable and
the dependent variable afterbotn have been adjusted for the effects of all other
independent variables in the equation thus far. Thus the pr statistic gives an
indication of the strength of the relationship between the aspects of an
independent variable and the dependent variable that are oniguo relative to the
other independent variables in the equation at that step. So in Table 1, the
portion of A80 score that remains after it has been adjusted for the effects of
both baseline MAACL-R depression score and exam outcome correlates with
the portion of Time 3 MAACL-R depression score that remains after it has also
been adjusted for the effects of both baseline MAACL-R depression score and
exam outcome at pr=.01.

The prz statistic for an independent variable is the median of the

remaining dependent variable variance, after it has been adjusted for all other

é ”C";

fiuwr'

I «ctr:
u- 'th ‘

tfificn
. v...

I N
lr.e:

61

independent variables in the equation, that is uniquely accounted for by an
independent variable after that independent variable has also been adjusted for
the other independent variables in the equation. It differs from the sr2 statistic in
that not only the independent variable, but also the dependent variable, is
adjusted for the effects of the other independent variables before the proportion
is calculated. Thus an independent variable’s pr2 represents the proportion of
the dependent variable variance that is not associated with the other
independent variables that is uniquely associated with the particular
independent variable. If only one independent variable is entered in the first step
of an equation, there are no other independent variables to partial out of the
dependent variable or the independent variable, and so in this special instance,
R2=sr2=pr2. Also, due to computational equality in the numerators of the
formulas for pr and sr (from which sr2 is calculated), for any given step, the F
value for sr2 and the t value for pr and pr2 are al_ways significantly different from
zero to the same degree (Cohen and Cohen, 1983, see p. 107). Therefore,
when only one independent variable is entered in a step of an equation, only
the F value is given and its significance applies equally to the sr2 and pr2 for
that set. When more than one independent variable is entered in a single step,
T values for each independent variable’s MM contributions are also given.
In using the analysis of partial variance in this study, the statistics of
interest are the pr and pr2 for the independent variable that is entered in the last
step of each equation. For example, in Table 1, a significant pr and pr2 value for

the A80 score by exam outcome interaction, if obtained, would have supported

I'UU

 

 

 

62

the hypothesis that this interaction predicted student’s depressive mood
responses to a unique and significant degree.

The significance or nonsignificance of the contribution of a covariate (in
this case, baseline MAACL-R depression score) to an analysis of partial
variance equation has no direct bearing on the interpretability of the results
pertaining to the research factors entered in subsequent steps of the equation
(Cohen and Cohen, 1983). By definition, the prz represents a research factor’s
relationship with the dependent variable after both have been adjusted for the
covariate, so its significance reflects its unique contribution regardless of the
significance of the covariate. However, other information can be derived from
the significance of a covariate in an analysis of partial variance, so such

information is provided for later interpretation.

 

W H‘: There will be en interegien between etedente’

41 ' in.| ‘nl: . m:. r:0 ”fro: th-ir mi -rm : m .nt : 0 cm:

 

. : r'ns :t- on thir xm tht will gran hir . :o. nt :n- rin-
r i r n

In order to test this hypothesis, an analysis of partial variance was done
to determine if Students’ ASQ scores interacted with their exam outcomes to
uniquely and significantly predict their MAACL-R depression scores at Time 3.
The results in Table 1 show that baseline MAACL-Fl depression score correlated
pr=.39 with Time 3 MAACL-R depression score and accounted for 16%
(sr"’=pr2=.16 p<.01) of the variance in Time 3 MAACL-R depression score.
Beyond that, neither ASQ score (pr=.01, ns) or exam outcome (pr=-.16, ns), or
their interaction (pr=.01, ns), correlated significantly with students’ enduring
(Time 3) depressive mood responses to the receipt of their exam grades. Thus,
this hypothesis was not supported.

Table 2 shows the corresponding findings (relative to Diathesis-Stress H‘)
for students’ immediate mood responses to receiving their grades. The A80
score by exam outcome interaction did not correlate uniquely with students’
immediate mood responses (pr=-.09, ns). However, the simultaneous entry of
A80 score and exam outcome in the second step yielded a significant
increment in the proportion of Time 2 MAACL-R depression score accounted for

by the equation (sr2=.19070, p<.05). Further, the within set analysis shows that

.muoonaSm ago I cansum .ouoom

do.v9I««
mc.vaI«
coumuouaoe ¢-Ao<<: N ends I oaauuuu> uccvcenon .quu

 

cauuouuoucu
because
do. eo.- om.e mm. dc. cu. aux» an om< n
ofi. co. on xemo.m oloouao none
00. co.- on ¢~.- vac
ed. es. om.m *oe.e ad. ea. huge» cannon N
M-Ao<<:
oo. oo. ue.a cc. co. co. ocufiouun fl
Awuav Auav up muouofivoun you new Asunv um» «a you c« you we
scuuuaouuoo couuedouuoo you-c«£uu3 «a =« you «a nu o>~u¢~5lsu Aavuouuuvoum huuco
Amuuueo aofiuuum wow u ucoaouucu ucoaouocu uo
vouusvm you b nacho

 

N candh

~0.vnI««

.Aoguom acunmounoav pouu>o&-unuq soozu o>uuoofip< uoouu<

canfiudcz any so ouoom I m-qu<<z .oufiuccouunoso udaum floccuusnuuuu< vouuquI osu co noaoouso
ucoao>ouzoo o>uuwwoc you meadow 5m zuudono~o one auuganuum o£u co nouooo uo omuuo>u I Omd .~ can
a momnmh you .muooqnam figu I o# sum .ououa couanounop ¢-Ao<<x n oauh I ounuuuo> unaccoaoo .«udm

 

acuuouuoucu
oaoouao
co. co. mm.s co. cc. a“. aux» so am< n
no. 3 . . on 84. 0333 5x...
oo. #9. on no. Om<
no. oH.- om.n am. No. ca. huuco cacao: N
. ¢-Ao<<:
ed. ac. dc ~ *«wn.s 0H. 0H. endgame: u
Auuav Auav up muouuavouu you you Annoy you «a you Cu you we
couuedouuoo couuodouuoo uom-:«£uu3 «a nu now «a cu o>~un~5lso AavuOuouvoum zuuco
unuuuua douuucm you u ucoaouocu acclouoca mo
voucsvm wow u novuo

13333

d Odadh

65
exam outcome correlated pr=.44 (p<.05) with Time 2 MAACL-R depression

score and uniquely accounted for 19% of the Time 2 MAACL-R depression
score variance (pr2=.19, p<.01). Therefore, exam outcome predicted students’s
immediate mood responses to the receipt of their exam grades, while ASQ

score and the A80 score by exam outcome interaction did not.

/

giantess-ems: H2: Eer etedente whe remive e lew exem grade, there
'I 9x: - 'nifi ..n .00 itiv orrI.-tin ntwnn h .11: f “rlt ”Vin :
our: onnl -ottri io.nl l t a lin- --ni ff" : in writ :n- r' o
r ivrn nttthr itfthirxm r

In order to test this hypothesis, an analysis of partial variance was done
using only data from students with a negative exam outcome to determine if
ASQ score uniquely predicted Time 3 MAACL-R depression score after both
had been adjusted for baseline MAACL-R depression score. Table 3 shows that
ASQ score was not correlated to Time 3 MAACL-R depression score for
negative exam outcome students (pr=.01, ns). Thus, this hypothesis was not
supported. Negative exam outcome students’ baseline MAACL-R depression
score correlated only pr=.11 with Time 3 MAACL-R depression score, and
uniquely accounted for only 1% (pr2=.01, ns) of the Time 3 MAACL-R
depression score variance. This is in contrast to the corresponding correlation
of pr=.40 and 16% of the variance accounted for by all students as shown in

Table 1. Table 4 shows that ASQ score also did not correlate significamly with

.oaooyao Edam o>uuuwoc m
nous muoofinam I canadm .oyoom coummoymmv M-Ao<<z N «Bub I ounouyu> ucdpcomoo .uuoz

 

yo. ou.- «H.N cu. Ho. Ho. 0m< a
m-Ao<<=
00. 4o.. mH.H no. oo. oo. ocuuomom H
Amyav Ayav up yum you as Anymv you «a you Cu you uo
couyouoyyoo couumuoyyoo cu ucoaoyOCu you «a cu o>uyousaso AuvyoyOupoym hyyco
uwuuyma Hduuymm you b uCoEoyo:H uo
onmavm yopyo

 

c ounce

.Aouuom coummoynoav pomu>0m-umua xoono

o>uuomfip< uoouu< ouauuudz any :0 oyoom I M-Au<<z .oywwccouumoao ouhym mcouuanuyyu<
pouuupoa may do mosoouso uCoEo>ou£oo o>uuo c: you mouoomnam uuuonoHu can

>yuuunmym any so moyoom uo owmyo>m I Om< .¢ one m mounds you .oaoou5o adxo o>uymwoc
a nuua myoonnzm I ouaadm .oyoom scummoyaov M-Au<<z m mafia I ounmuyc> unaccoaoa .Mwma

 

00. Ho. cu.~ oo. oo. #0. Om< N
. . M-Ao<<z
Ho. uu. mu u mu. Ho. Ho. ocuuomun H
Auyav Ayav up uom you «a Anymv you «a you Cu you uo
couywuoyyoo coyuwuoyyoo cu uCoEoyoCu you «m cu o>uuouaa=u Amvyoyouvoym xyuco
uduuymm Hmuuymm you m unoaoyoCH uo
poymscm yopyo

 

n oypay

67

Time 2 MAACL-R depression score for students with negative exam outcomes

(pr=-.10, ns).

'31 : i- r. 3' l’ t 0-nt who : 51V: - hih -x.m or.” h--l‘
will bet E e eignifieent beeitive er negetive eerrelaien between the flete bf
‘11 0.: . h'ovin - our: 00,:r‘ll t‘trio i-.nl l = .0- ”lin .no in r:: : in

11:11 = 0 l" 0 our: iv mu- 0 :I nt ,0 th- r- =l.-_ f hil‘
5.999.5-

In order to test this hypothesis, an analysis equivalent to that described
for Diathesis-Stress H2 was done using only the data from students with a
positive exam outcome. Table 5 shows that, as was the case for negative exam
outcome students, positive exam outcome students’ ASQ scores did not
correlate with their Time 3 MAACL-R depression scores (pr=-.01, ns). Thus, this
hypothesis was supported. Baseline MAACL-R depression score correlated
pr=.52 (p<.05) with Time 3 MAACL-R depression score, and uniquely
accounted for 27% (pr2=.27, p<.01) of the Time 3 MAACL-R depression score
variance. This is in contrast to the corresponding correlation of pr=.11 and 1%
of the variance accounted for by students with negative exam outcomes as
shown in Table 3. Table 6 shows that ASQ score did not correlate significantly '
with Time 2 MAACL-R depression score for students with positive exam

outcomes (pr=.16, ns).

. .oaooyso auxo o>uyuuoa u
nyu3 uyoomnsu I ounaum .oyoou couuuoyaov x-ao<<= N oaua I ounoyyu> ycopconoa .Mmmm

 

no. my. m~.~ as. No. No. am< N
M-AU<<=
no. um. ¢~.H «H.u nc. no. ocuuouun H
a~yav Aynv uv you you «a Auyuv you «a you cu you uo
couyouoyyoo acyyouoyyoo nu yCoEoyocu you «a cu o>yyuuza=o Auvyoyouvoym hyyco
youyyon uuyyyum you b yCoaoyoGH uo
voyusum yovyo

 

n uyuoayo : uuoyyu-u uozyuy oy ouyyu o

w ouauh

yo.veu**

.Aouuom coyuuoyaoov vouy>om-yuu4

xoono o>uyoofiv< yoouu< ouayyuaz ony :o oyoou I x-Ao<<z .oyyuccouyuoac oumym
Mucouynnuyyy< vouuyvoa ony co uoaooydo yCoao>oynoo o>yyuwoc you uouuounsu hyuuunouo

van yyuunuym ony co uoyoou uo omuyo>o I Om< .m paw m uounus you .oaooyso auxo o>uyyuoa
u :yya uyoomnsu I ouaaom .oyoou aouuuoynov M-Ao<<z m oauh I ounouyu> ycuvconoa .mmma

 

00. Ho.- m~.N oo. oo. nu. 0m< N
. M-Ao<<z
mm. mm. ¢N.H oumm.w mm. mm. ocuuouom H
Auyav Aymv up you you um Auyuv you «a you nu you uo
coyyuuoyyoo couyuuoyyoo Cu yCoEoyUCu you «a nu o>uyuuaa=o Auvyoyouvoym hyyco
uofiyyun Houyymm you m yCoaoyoCH uo
coyoavm yocyo

 

n udauh

69
Tt fth lm itinh th IM itinI-I“5

Qeueel Medietien H‘: Fer etedente whe reeeive e lbw exem grede, there

will , ; . ionifi 90 itiv --.-rrltin . ‘twnzz . nt ’ . r'o i-.nl I
m r frthirmi rmxmnthirr iv ttri infrthir
mrfermeng‘e en their migflerm exem.

In order to test this hypothesis, a standard multiple
regression/correlation analysis (there are no relevant covariates to justify an
analysis of partial variance) using only data from students’ with a negative exam
outcome was done to determine if ASQ score accounted for a significant
proportion of the PAC score variance. Table 7 shows that ASQ score correlated
pr=.68 (p<.01) with PAQ score and accounted for 46% (sr2=.46, p<.01) of the
PAQ score variance for these students. Thus this hypothesis was supported.
However, for students with a positive exam outcome, ASQ score also correlated
pr=.46 (p<.05) with PAQ score and accounted for 21% (sr2=.21) of the PAQ
score variance, as shown in Table 8. Also, for all students in the research
sample, ASQ score correlated pr=.33 (p<.05) with PAQ score and accounted

for 11% (pr2=.11, p<.05) of the PAQ score variance, as shown in Table 9.

7T)

wc.vnlo
.uyuonyzu gun I ounaum .oyoou c<m I ouauuyu> ycuvcoaoa .umuz

 

 

. 2. mm. 3; «a: S. 3. 02 y
Anynv Aynv uv you you «a anyuv you «a you :4 you uo
scuyuuoyyoo :Ouyuuoyyoo cu ycoaoyocu you «a :u o>uyuuslfio AuvyOyuuvoym myyco
yuuyyun uuuyyom you m ycoloyocu uo
voyusam . yovyO
a O ..s.. .HytH—vo Inn... .0 I .1 :3. . ..L , , . . .l .. ,. I” ,. v - ,_.
a oyyus
no.vnlo
.oEooyso :uxo o>uyyuoa u :yu3 uyooHASu I ounaum .oyoou a<u I ounuuyu> ycuvcoaon .mudm
um. 04. e~.y *cm.u um. um. Om< u
Ayyav Aynv uv you you Nu Auyuv you a: you Cy you uo
scuyuuoyyoo coyyuyoyyou cu ycoaoyocu you «a nu o>uyuuaasu Auvyoyouvoym uyyco
yuyyyon youyyum you a ycoaoyozu uo
voyusum yovyo
.— u «059.3. y. 01.60“ dual 00 «.u 4.. . .l«. 4. 1:. -. l .. . . . , . -ru . Orig:
a ounce
yo.valoo

.uolooyao luxo yuo£y you

ucoyyanuyyyo .uyooa :u voyauuoa sous: oyuuccouyuoso ucouyayuyyy< yousouyyum ocy
you uouuounsu xyuuunouo pcu yuuunuym ozy co uoyoou uo omuyo>u I can .oyuuCGOuyuoso
ouxym Hucoyysnuyyy< voyuuvoa ozy co uoeooyao ycolo>ou£uu o>uyumoc you uouuounsu
xyuuunoyo vcu xyyuuyoym ocy co uoyoou uo omuyo>u I Om< .o can a .u uouauh you
.oaooyso :mxo o>uyuwoc u Syua uyooH£Su I ouaEum .oyouu O<m I ounuuyu> unneconoo .uumz

 

we. we. mu.~ «oum.- we. cu. Om< u
Anyav Ayav uv you you «a Auyuv you «a you cu you up
couyuuoyyoo coyyuuoyyoo :y ycoeoyucu you «a nu o>uyoualsu Auvyoyouvoym xyyco

Huuyyua yuuyyum you a ycoaoyocu uo
voyucvm yovyo

aIudmuammaHuIdm4uuduuuIAdauquIuuIuuqHluuuluuduuquluqlnuullau
h Gun-H

71
Qeueel Medietien H2; Fer etugeme whe receive a lbw exem grede, there

will .0: = inifi ,nt .00 itiv o-.-rrltin oztwn:= o-nt ' r- 9: iv .0ttri ion
or pir urf- mn : nthir mi rm -.xm .no unt ’ :n- ri o. ._;. : i :
m r n
In order to test this hypothesis, an analysis of partial variance was done

using only data from students with a negative exam outcome to determine if
PAQ score uniquely predicted Time 3 MAACL-R depression score after both
had been adjusted for baseline MAACL-R depression score. Table 10 shows
that PAQ score did not significantly correlate (pr=-.12, ns) with Time 3 MAACL-
R depression score for these students. Thus, this hypothesis was not
supported. However, Table 11 shows that PAQ score correlated pr=-.53

(p< .05) with Time 2 MAACL-R depression score and uniquely accounted for
over 23% (pr2=.283212. p< .05) of the Time 2 MAACL-R depression score

variance.

I um: in H3: F _-_: who ram : . low : an . se,: mr:
wil u: . inifi nt H itiv .orrl:. ion u: ::n 0.: ’r- 9301N3 =1 ri 'n
for -ir uzr‘f- n = on hir miurm .xm = o 0:19» :no ri o 0.=- = ° =
III-n. : MI : tti Mum-mt f n 01-: an -:tw:: t 0:1.
=.Iul.0 '-.nl ml: .. m:. -r:- with thir min-rm : m :n-thir :n- rm
m r n
In order to test this hypothesis, an analysis of partial variance using only

data from students with a negative exam outcome was done to determine if

72

I . no.valu
.oaooyao auxo o>Hyowoc
u AyHa uyoofinsu I oHaaum .oyoou CoHuuoyaov &-Ho<<= N oaHH I oHAtuu> ycuvcomoo .mmma

 

mm. mm.- ey.~ emm.m my. «N. c<m N
m-uu<<z
oo. eo.- my.y mo. oo. oo. oeyyeeen y
ANyav Hyav uU you you «a Huyuv you «a you :H you uo
coHyoHoyyoo coHyuHoyyoo :H yCoaoyocH you «a CH o>HyuHaasu AuvyoyoHonm hyyco
Htuyua Htuyum you u yzoaoyocH uo
voyuscm yovyo

 

HH oHnua

.uoaooyao auxo yuony you uncuy=AHyyyu .uyoonnau voyauuoa

zouna oyuoccoyyuoao ucoHyanuyyy< yuHaoHyyum o£y you uoHuoundu hyHHquHo can hyHHHnuym
osy co uoyoou uo owuyo>u I O<m .HoHuom :oHuuoynoav vouH>oM-yuHH goose o>Hyoo pc
yoouu< oHa stz ony :o oyoou I m-qo<<z .HH vcu oH uoHnuH you .oaooyso Bozo o>Hyu o:

u nyHa uyoo you I oHasum .oyoou :oHuuoyaov m-Ho<<z m oEHH I oHntuu> ycuvconoo .mmmz

 

Ho. NH.- dH.N . HN. Ho. no. Q<m N
. m-qo<<=
Ho. HH. mH H «H. Ho. Ho. oGHHouum H
Huyav Hyav uv you you «a Huyuv you «M you :H you uo
:oHyuHoyyoo :oHyuHoyyou CH ycoaoyocH you «a cu o>HyuHaado AuvyoyoHvoym hyyco
Htuyua Htuyum you m yCoEoyocH uo
voyosvm yovyo

 

oy eyuey

73
PAQ score uniquely predicted Time 3 MAACL-R depression score after both

had been adjusted for baseline MAACL-R depression score ebb ASQ score.
Table 12 shows that PAQ score did not correlate significantly (pr=-.18, ns) with
Time 3 MAACL-R depression score, and uniquely accounted for only 3%
(pr2=.03, ns) of the Time 3 MAACL-R depression score variance beyond the
0.0% (pr2=.00, ns) uniquely accounted for by ASQ score. Thus, this hypothesis
was not supported. However, Table 13 shows that PAQ score correlated pr=-
.65 (p< .01) with Time 2 MAACL-R depression score and uniquely accounted for
43% (pr2=.43, p< .01) of the Time 2 MAACL-R depression score variance

beyond the negligible 1% (pr2=.01, ns) accounted for by ASQ score.20

Qeeeel Medietien H", There will be e eignflieent pQS'EiVQ eerreletjen

u: ::n unt’f-olin oth-u: n me r:-tw- .- - r h
r itfthirxmr n t nt’nrin r ivm
1.950.005.95-

In order to test this hypothesis, an analysis of partial variance was done
using only data from students with a negative exam outcome to determine if
Time 3 hopelessness uniquely predicted Time 3 MAACL-R depression score
after both had been adjusted for baseline MAACL-R depression score.21 Table
14 shows that Time 3 hopelessness did not correlate significantly (pr=-.19, ns)
with Time 3 MAACL-R depression score for these students. Thus, this

hypothesis was not supported. Similarly, Table 15 shows that Time 3

74

Hc.vnluu
.oaooyso auxo o>Hyuwoc

u zyH3 uyoonndu I oHaBum .oyouu :oHuuoynov m-Ho<<z N oaHH I oHntuu> unaccoaoa .mmmz

 

mu. me.- MH.m «*No.m Ne. nu. d<m n
Ho. oH.- «H.N «H. Ho. Ho. dm< N
M-Ho<<z

oo. «o.. mH.H no. co. co. ocHHouun H
HNyav Ayav up you you Huyuv you «a you 6H you uo
coHyuHoyyoo soHyuHoyyoo Nu :H you «a nu o>HyuHaaso HuvyoyOHvoym hyyCo

Htuyua Htuyum ycoaoyocH ycoaoyocH uo
voyusvm you m yovyo
n uuuofiuozéu anoHvo: Han do . y .a HQ. _. .. . u H. . .. - . . , y I. 4. Ian. .

nH oHyua

.uoaooydo auxo yHony you ucouyanyyyu .uyuonnau
voyauuoa noH£3 oyHuccoHyuoso ucoHyanyyy< yuHaoHyyum ony you uoHuounau hyHHunoHo can
hyHHHnuym ony co uoyoou uo owuyo>u I O<m vcu .AoHuom :oHuuoyaoav vouH>om- yuHH xoono
o>Hyoonv< yoouu< oHaHstz ozy co oyoou I a- Ho<<z .oyHucaoHyuodo loym HucoHyanyyyc
voHquoa ony co uoaooyao ycoao>oynou o>Hyuwoc you uoHoounsu AyHHunoHo vcu xyHH nuym
osy co uoyoou uo omoyo>u I Om<. MH use NH uoHnuH you .oaooyso auxo o>Hyu on u

nsz uyoomnnu I oHaaum .oyoou :oHuuoynoo a Ho<<z m oaHH I oHaoHyu> ycupCoaoa .wudz

 

no. mH.- MHHm Nu. no. uc. O<m m
8. 8. 3 N , oo. oo. S. 03 N
. ¢-Hu<<z

Ho. HH. mH H wH. Ho. Ho. ocHHouun H
Auynv Hynv uc you you Hayuv you Na you 6H you uo
:oHyuHoyyoo :oHyuHoyyoo Na cH you «a :H o>HyuHaado HuvyoyoHvoyu hyyco

Htuyua Htuyum ycoEoyocH ycoaoyocH uo
voyusum you b yovyo

u o y u

NH OHAIB

75

.oaooy:o auxo o>Hyuwoc u
nyH3 uyoonnzu I oHuaum .oyoou coHuuoyuov &-Ho<<= N oEHH I oHntuu> ycuvcouoa .uqmz

 

uuoCuuoHouos
oo. mo. mH.N no. oo. co. m oaHH N
M-Ho<<z
oo. mo.- «H.H Ho. co. co. ocHHouun H
Huyuv Hyuv uv you you NM HNyuv you Nu you CH you up
:oHyuHoyyoo :oHyoHoyyoo cH ycoaoyocH you NM :H o>HyuHaaao AuvyoyoHvoyu uyyco
HoHyyuu Htuyou you u ycosoyocH uo
voyuaum yocyo
e uHuoayo...M oHyquo: uuauo .y do,. u.a.o .; . u- . . -u . .0 ;;;

nH OHAuB

.oHuom uuofiuuoHouo:
xoom ony co oyoou I uuoCuuoHouon vcu .HoHuum :oHuuoyuoov vouH>om-yuHH xoono o>Hyoonv<
yoouu< oHtu a: ony co oyoou I u-Ho<<= .mH ncu «H uoHAuH you .oaooyao auxo o>Hyuwoc u

sst uyoo nzu I oHuEum .oyoou :oHuuoyuov «IHU<<Z m oBHH I oHatuu> ycuUCouon .mqmz

 

uuoCuuoHouoz
uo. oH.- mH.N . we. co. no. N oEHH N
m-Hu<<z
Ho. NH. uH.H HN. Ho. Ho. ocHHouum H
Anyuv Hyuv uv you you «x Auyuv you Nu you :H you uo
coHyuHoyyoo coHyuHoyyoo CH ycoaoyocH you Na 6H o>HyuHsadu AuvyoyoHvoyu uyyco
Htuyou Htuyuu you u ycoaoyocH uo

uoyusvm

 

«H OHHIB

76
hopelessness did not correlate significantly (pr=.05, ns) with Time 2 MAACL-R

depression score variance.

Qeueel Medietien H5: There will be e eignifieent peeitive berreletien
0:: n t o-nt ’ in I'=. : in fu-Iin f iii-9:: n- from 0- ”Iin Im-1

:: mth-o o:I-w to tw- ca ”.ftrth r it of xm or.” .n- t --nt’
n rin r iv m r n
In order to test this hypothesis, an analysis of partial variance was done

using only data from students with a negative exam outcome to determine if
their net change in hopelessness (Time 3 hopelessness - baseline
hopelessness = net change in hopelessness) uniquely predicted Time 3
MAACL-R depression score after both had been adjusted for baseline MAACL-R
depression score.” Table 16 shows that net change in hopelessness did not
correlate significantly (pr=.02, ns) with Time 3 MAACL-B depression score for
these students. Thus, this hypothesis was not supported. Table 17 shows that
net change in hopelessness correlated nonsignificantly at pr=-.34 (p=.20) with
Time 2 MAACL-Ft depression score and uniquely accounted for over 12%

(pr2=.121925, ns; p=.2021) of the Time 2 MAACL-R depression score variance.

While the above are the results of specific tests of each of the

hypotheses in this study, the interpretation of those tests can be facilitated by

viewing a concise and meaningfully organized presentation of the means for

77

.oaooyao auxo o>Hyuwoc u
nyHa uyoonndu I oHusum .oyoou :oHuuoyuov M-Ho<<= N oaHH I oHntuu> ycuocouoa .mmmz

 

uuocuuoHouon
Ny. mm.- my.N ye.y Ny. Ny. :« emeeno uez N
m-Ho<<z
co. No.- ey.y yo. co. co. oayyeeem y
HNyuv Hyuv up you you NM HNyuv you am you :H you up
:oHyuHoyyoo coHyuHoyyoo cH ycoaoyocH you Nu :H o>HyuHsaa Huvyoyoyvoyu uyyco
HoHyyou Htuyuu you u ycoEoyocH . uo
voyuaum yovyo
uHuonyo...: COHyquo: Huuauo .y ctuulé. - 9..-- . I- ,. .. .au. ., o u. .

NH OHQGB

.m oeHH oy oEHH soyu oHuom uuocuuoHouoz Hoon ony co oyoou CH oocoyou Hv
I uuocuuoHouo: :H o coco yo: can .AoHuom :oHuuoyuoov vouH>om-yuHH xoozo o>Hyoo v<
yoouu< oHtu a: osy co oyoou I m-qo<<z .NH can eH uoHnuH you .oaooyao auxo o>Hyuwoc u
nyHa uyoo nsu I oHuEum .oyoou coHuuoyuoc M-Ho<<z n oBHH I oHntuu> ycovCouoo .mmmm

 

uuocuuoHouoz
co. No. NH.N . yo. co. No. cu emcene uez N
. . M-Ho<<z
Ho. NH. «H H HN. Ho. Ho. ocHHouum H
ANyuv Hyuv uv you you Na ANyuv you Na you cH you uo
coHyuHoyyoo :oHyoHoyyoo :H ycoEoyocH you NM :H o>HyuHaadu HuvyoyOHvoyu hyyCo
Htuyuu Htuyuu you u ycoaoyocH uo
voyusvm youyo

 

uH OHauH

78

important variables in this study. Table 18 shows the baseline, Time 2, and Time
3 MAACL-R depression score group means for the total research sample, and
for the positive and negative exam outcome groups. It also shows each group’s
mean ASQ score as measured at baseline and PAQ score as measured at Time
2. Table 18 shows that, at baseline, the mean MAACL-R depression score and
mean ASQ score for each portion of the sample appear to be about equivalent.
Students with a negative exam outcome showed an elevated immediate
depressive mood response to the receipt of their exam grades (mean Time 2
MAACL-Ft depression score=66.941), and made slightly less global and stable
attributions for their exam performance (mean PAQ score=2.534) than would
have been expected based on their mean baseline ASQ score (3.206), but their
mean MAACL-R depression score dropped to below baseline two days later
(Time 3 MAACL-R depression score=49.674). Students with a positive exam
outcome showed a reduced immediate depressive mood response to the
receipt of their exam grades (mean Time 2 MAACL-R depression
score=46.500), and made more global and stable attributions for their exam
performance (mean PAQ score=4.635) than would have been expected based
on their mean baseline ASQ score (3.206), but their mean MAACL-R depression
score returned to about baseline two days later (Time 3 MAACL-R depression
score=53.923).

The mean Time 3 hopelessness scores were 1.707 for the total research
sample, 1.520 for students with a negative exam outcome, and 2.000 for

students with a positive exam outcome. The mean net change in hopelessness

79

table 18
W

 

 

 

 

 

Time 1 Time 2 Time 3
Portion (immediately after (2 days after
of (baseline) receiving grade) Time 2)
sample
MAACL-R ASQ MAACL-R PAQ -HAACL-R
Total
research $5.60 3.07 54.58 3.80 52.23
sample
(n-43) sd-l4.78 sd-.96 sd-23.24 sd-l.68 sd-l6.23
Students
with a 56.26 3.21 66.94 2.53 49.67
negative
exam
outcome sd-l4.69 sd-.76 sd-33.82 sd-l.12 sd-8.lo
(n-l7)
Students
with a 55.16 2.97 46.50 4.64 53.92
positive
exam a
outcome sd-15.11 sd-l.07 sd-l.82 sd-l.45 sd-l9.83
(n-26)

 

 

 

 

 

 

 

 

E931: MAACL-R - Multiple hrfect Adjective Check List-Revised: depression
score. ASQ - Attributional Style Questionnaire: score for achievement
events. PAQ 9 Particular Attribution Questionnaire: attribution for exam
outcome. SD - Standard deviation.

- 23 of 26 bad a score or 47.

80

scores was -.585 for the total research sample, -.520 for students with a
negative exam outcome, and -.677 for students with a positive exam outcome.
These reflect values and changes in values on a 20 point scale.

A correlation matrix depicting the relationships between the independent
and dependent variables of this study for the full research sample is presented
in Table B, Appendix L. Appendix L also shows a corresponding correlation
matrix for the subset of the research sample that had a negative exam outcome
in Table C, and another for the subset that had a positive exam outcome in
Table D. Appendix L also contains tables depicting corresponding correlation
matrices for the full subject sample (as opposed to the selected research

sample) in Table E, and for various meaningful subsets of that sample in Tables

F-K).

DISCUSSION

The purpose of this study was to conduct a test of the hopelessness
theory of depression (HTD), and in particular, tests of the Diathesis-Stress
component and Causal Mediation component of HTD. While various tests of
HTD and the revised learned helplessness theory of depression (LHTD-R) have
been conducted (for reviews, see Barnett and Gotlib, 1988; Brewin, 1985;
Coyne and Gotlib, 1983; Peterson and Seligman, 1984; and Sweeny et al.,
1986), these tests have often disregarded important theoretical components of
HTD in a manner that severely limited the degree to which they could justifiably
refute er lend support to HTD (Abramson et al., 1986; Abramson et al., 1988;
Alloy et al., 1988). That is, while most tests of HT D and LHTD-R have
adequately controlled for subjects’ baseline conditions of having a
depressogenic attributional style, none to date have also adequately controlled
for subjects’ baseline conditions of experiencing important negative life events
(INLE) or incidence of non-hopelessness subtypes of depression. WlthOU‘t
controls for these factors, eny set of results pertaining to the relationship
between the incidence of having a depressogenic attributional style and the
incidence of depression, whether supportive or refuting, can be explained by

HT 0, thus rendering the theory untested. Therefore, in this study all subjects

81

82

who reported experiencing an extremely negative life event in the six months
prior to the experiment or who scored higher than nine on the Beck Depression
Inventory (BDI) were removed from the sample. Then a time series study was
done measuring students’ baseline levels of depressogenic attributional style,
hopelessness and depressive mood for comparison with their immediate (Time
2) and enduring (Time 3) mood responses following the receipt of their midterm
exam grades. A measure of the attributions students made for their
performance on their exam was made directly after their Time 2 mood response
was assessed, and a second measure of hopelessness was made after their
Time 3 mood response was assessed.

It will be considered below that the results of this study raise questions
about the practical versus heuristic utility of examining HTD as a complete and
resolved theory, and even about the overall validity of HTD. It will be shown that
this study did not support a pattern of hypotheses which, when considered
together, could have provided strong support for HTD. When considering
possible explanations for these data, it will be recognized that the results of this
study are consistent with, although not conclusively supportive of, trait theory
for depression and alternative conceptualizations of the relationship between
cognition and depression (e.g., the contention of Barnett and Gotlib, 1988, and
Lewisohn et al., 1981, that attributional style is concomitant with depression). It
will be further recognized that limitations in the ability of this study to isolate a
subject sample demonstrating an enduring depressive mood response to an

important negative life event may have contributed to its inability to conclusively

83

evaluate HTD. Methodological issues raised by this study and recent
developments in the HTD literature will then be discussed. Finally, when this
study is considered in the context of trying to test a rapidly and continuously
evolving theory, it will be suggested that the specific limitations of this study are
overshadowed by the more apparent implication that it is premature to attempt
to evaluate HTD as though it were a complete and resolved theory. As a
foundation for later elaboration on these points, a discussion of the hypothesis
testing of this study will now be presented.

Six of the eight hypotheses of this study focused on predicting subjects’
Time 3 depressive mood responses to an INLE Ge, a negative exam outcome).
This focus was chosen because it has been suggested that jmmeeiete mood
responses to a negative exam outcome can be transitory in nature and
attribution-independent for some students, while a measure of depressogenic
attributional style has been able to predict ebebmg depressive responses to a
negative exam outcome in other students (Metalsky et al., 1987). However,
Table 18 shows that the mean depression scale scores on the revised Multiple
Affect Adjective Check List (MAACL-Ft) for both positive and negative exam
outcome groups were actually slightly lower at Time 3 than they were at
baseline. Similarly, baseline MAACL-R depression score correlated pr=.40
(p<.01) with Time 3 MAACL-R depression score, suggesting that the
depressive mood of these students at Time 3 was similar to their mood at

baseline. Therefore, as the hypothesis testing is summarized below, the

84
implications of the results of this study for HTD will be qualified in light of this

absence of an enduring mood response.

Wlth regard to the Diathesis—Stress component of HTD, this study did not
support the assertion that students’ attributional styles would interact with their
exam outcomes to predict their subsequent enduring mood responses
(Diathesis-Stress H‘). Rather, the interaction of these factors did not correlate
significantly (pr=.01, ns) with Time 3 MAACL-R depression scores. In light of
this, the inability to show that attributional style predicted Time 3 MAACL-R
depression scores for those with negative exam outcomes (Diathesis-Stress H2)
is not surprising, and the success in showing that attributional style did not
predict Time 3 MAACL-R depression scores for those with positive exam
outcomes (Diathesis-Stress H3) lacks relevance. The purpose of Diathesis-
Stress H2 and H3 was to confirm that, if the interaction predicted in Diathesis-
Stress H1 had occurred, it occurred in a fashion consistent with the Diathesis-
Stress component of HTD. Specifically, these hypotheses were to show that the
nature of the interaction would be for a high Attributional Style Questionnaire
(ASQ) measurement of students’ attributional styles toward making global and
stable attributions for negative achievement events to predict an 111119868 in
MAACL-Fl depression score from baseline to Time 3 for students with a
negefue exam outcome. Therefore, the inability to support Diathesis-Stress H2
can be credited to the absence of an ASQ score by exam outcome interaction
upon which to elaborate, and the absence of a Time 3 MAACL-R depression

score effect to predict. The success of Diathesis-Stress H3 appears to be due to

85

the hypothesis being phrased in a manner that requires a significant result to
accept the null hypothesis. Therefore, this hypothesis could be rejected only if a
high ASQ score significantly predicted an M in MAACL-R depression
score from baseline to Time 3 for students with a peefiye exam outcome.
Clearly then, Diathesis-Stress H3 can hold meaning relevant to the Diathesis-
Stress component of HTD only if elaborating on significant results supporting
Diathesis-Stress H1 and H2.

Wlth regard to the Causal Mediation component of HTD, it was shown
that students’ attributional styles as measured at baseline uniquely accounted
for over 45% of the variance in students’ subsequent attributions for their
performance on their exam (Particular Attribution Questionnaire; PAQ) for
students with a negative exam outcome (Causal Mediation H‘). However, the
assertion that students’ PAQ score would predict their Time 3 MAACL-Fl
depression scores was not supported (Causal Mediation H2). Correspondingly
then, there was also no support for the assertion that the relationship between
PAQ score and Time 3 MAACL-Fl depression score would still be present after
both were adjusted for students’ ASQ scores (Causal Mediation H3). Similarly,
there was no support for the assertions that students’ Time 3 hopelessness
scores (Causal Mediation H‘) and their net change in hopelessness from
baseline to Time 3 (Causal Mediation H5) would predict their enduring
depressive mood responses.

The first three hypotheses for the Causal Mediation component of HTD

were designed to isolate and test specific components of the etiological chain

86
proposed by HTD. Support for Causal Mediation H1 did not hinge upon the

occurrence of an enduring depressive mood response. According to the Causal
Mediation component of HTD, given an INLE, the formation of causal
attributions is expected to precede a hopelessness type depressive mood
response, and such a response is only expected if the causal attributions are
depressogenic in nature. So the established relationship between ASQ score
and PAQ score in Causal Mediation H1 need not be qualified by the absence of
a Time 3 MAACL-R depression score effect. On the other hand, given no
enduring depressive mood response to predict, the inability to find support for
Causal Mediation H2'5 is not surprising.

Clearly, these data do not lend support to HTD. Toward understanding
these unexpected findings, it is important to consider how to characterize best
the Time 3 MAACL-R depression scores that were obtained. Most hypotheses
focused on how to predict an enduring mood response to an INLE. Without
such a response, there was nothing to predict, and many hypotheses were left
unsupported. At issue is first, whether the Time 3 results are valid, and second,
how the students’ moods at Time 3 are to be explained.

Regarding the validity of these results, it is worth noting that students
Mime mood responses to their exam outcomes were very similar to those
of the students in the Metalsky et al. (1987) study. Tables 19 and 20 show the
results of both studies for a test of the assertion that students’ attributional
styles would interact with their exam outcomes to predict their subsequent

immediate (Time 2) mood responses.23 Metalsky et al. (1987) found that only

EI7

 

ao.vulu
.uyoonasu HHu I oHusom .oyoou coHuuoyuoo ¢.Ho<<: N oeuh I oHneHyu> yseocouoa .uuuz
:oHyoeyoysH
osooyso euxo
ye. ee.- an.« an. yo. eN. can. an emu n
aH. «a. on euno.n osooyao sexo
co. ec.. on eN.. au<
ey. es. an.n ecu.e ey. ey. huge. «Hosea N
m-ao<<:
co. co. . He.H oo. oo. oo. ocuHouem H
Huyuu Ayuu uu uyoyuuooyu you you Huyuu you «a you cu you uo
coHyuHoyyoo coHyeHoyyoo you-cy£st «a cu you «a cu o>yyuHaeso AuuyoyuHooyu uyyso
Huyyyou Hoyyyou you y ycosoyocu ycosoyocH uo
ooyucom you u yooyo
. . u . .. . . .. u . . .34. . . I. ._ 3;“
udduuwuddddaHdaaIaHIauudHdaIauINIdldHIuauluuludqulqaulualuudndddquluudllau
0N eHaeh
Hoo.vulosu
.HoHeom couuuoyuoou oouu>oz.yuya xuozo o>uyoono< yuouu< oHtuHat ogy co oyoou I x.au<<:
.oyyoccouyuoco oHuym Hucouyabyyyy< ooyuu os osy co uosooyoo ycoso>ou£os o>Hyemoc you uoHuuunsu

uyuHuboHo ecu uyyuubuym ozy co uoyouu uo o uyo>u I Om< .9 use aH uoHyuh you .uyoonasu HHe I ownssm

.oyouu couuuoyuoo m-ao<<: N osuh I oHaeyye> ycoocouoa .Hoan. .NouHu .Hu yo uquuyo: soyu oHAuh .

 

soHyouyoycH
.. osooyso
ye. No.. ee.y ye. cc. NN. aux. ye om< n
oH. nu. ca uean.u olooyao lexo
No. nH. om nN.H cu<
eN. so. a. N ...ee.yy ey. NN. anon. eyesee N
m.ao<<z
no. oH. No.H co.n no. no. ecuHouon H
Anyuv Ayuv uu uydyuuuoyu you you HNyuv you «a you cu you uo
coyyuHoyyou couyoHoyyoo you-cu:yus Na cu you Nu :— e>HyuHssso Huuyoyouooyu uyyco
Hoyyyuu Houyyou you y ycosoyucu ycoeoyucu uo
voyeavm you u yooyo
HIuHuuauuaHaIuuauuuuuHuuaauqarqulauuquuu-uulurqlqu
_ . u . . H31. .4 u A . O . _ _ l . . . ' .

9H eHaeh

88
exam outcome correlated significantly (pr=.43, p<.001) with Time 2 MAACL-R

depression score variance (see Table 19). Similarly, this study found that only
exam outcome correlated significantly (pr=.44, p<.05) with Time 2 MAACL-R
depression score (see Table 20). Both studies found that baseline MAACL-R
depression scores did not uniquely predict Time 2 MAACL-R depression scores
and that the A80 scores by exam outcome interaction accounted for virtually
0.0% of the Time 2 MAACL-R depression score variance. The clear similarity
between these findings suggests that, at least with regard to students’
immedjete responses to the receipt of their exam grades, the subjects and the
measures of the two studies functioned comparably.

One explanation for the obtained results is that depression may be
primarily a trait dependent phenomenon, and that all those prone to showing
depressive responses to negative events were selected out of the research
sample. This would suggest that the processes that comprise the etiological
chain proposed by HTD are not real and therefore could not be revealed by this
study. Another explanation is that the absence of an identifiable enduring
depressive mood response in the research sample was due to the subjects not
experiencing the return of their exam grades as an INLE. This would suggest
that without students showing an enduring depressive mood response, several
hypotheses in this study could not be tested. Each explanation will be
considered respectively.

If attributional style does not interact with lNLEs to cause depressogenic

attributions and thus cause hopelessness and then depression, the results of

89
this study might be reflecting such a reality. This argument would be consistent

with the position of Barnett and Gotlib (1988, p. 106) that ”there are a number
of reasons to expect negative results in research... [that investigates] the
interaction of attributional style and negative life events... [with] prospective
research using proper measures of depression” In support of this expectation,
they point to two issues. First, Barnett and Gotlib (1988, p. 106) state that “our
review of the literature suggests that an abnormal attributional style is not
characteristic of the cognitive functioning of either premorbid or remitted
depressives.“ Therefore they suggest that researchers will be unable to identify
such a cognitive vulnerability in a prospective study that conforms to their
design recommendations. Second, in response to some reports of low reliability
for the A80, Barnett and Gotlib (1988, p. 106) state that:
The low reliability of the measure may account for the general lack of
significant results, due to the attenuation of all correlations involving the
ASQ. A different and more substantive interpretation, however, is that the
low reliability of the A80 is not purely a psychometric problem; rather, it
may accurately reflect the lack of cross-situational consistency in
subjects’ causal attributions (Cutrona,[Russell & Jones], 1985; Miller,
[Klee & Norman], 1982). This interpretation is consistent with the paucity
of empirical evidence of a trait-like cognitive vulnerability to depression
and suggests the need to revise the reformulated learned helplessness

model of depression.

90

The inconsistency between the results of this study and those of
Metalsky et al. (1987), per se, does not challenge the possibility that this study
may accurately reflect a reality that would be disconfirming of HTD. Given the
cogent discussion of the implications of research methodology by Barnett and
Gotlib (1988, see pages 36-38 above) it could be argued that the students who
showed an enduring depressive mood response in the study of Metalsky et al.
(1987) may have already been depressed, and that their attributional style
interacted with their receipt of a negative exam outcome to exacerbate their
depressive condition. That is, students who are already depressed could be
viewed as exhibiting depressive symptoms in a trait-like fashion, including
depressive attributions, which would leave them vulnerable to an exacerbation
of their depression. Similarly, it could be suggested that since depressed
students were selected out of the sample for this study, there were no students
exhibiting trait-like depressive attributions, leaving no opportunity for such an
effect to take place in the remaining students. Further, it could be suggested
that the lack of an enduring depressive mood response in the remaining
students argues for the contention of Gotlib and Barnett (1988) and Lewinsohn
et al. (1981) that attributional style is only concomitant with depression, and
may at that time act to exacerbate the condition, but is not a characteristic that
predisposes a never-depressed or fully remitted depressive to a subsequent
depression. This argument is consistent with the notable similarity in students’
immediate mood responses to their exam outcomes between subjects in this

study and those of the Metalsky et al. (1987) study (see Tables 19 and 20, p.

91
88). These results suggest that for this study there was a sufficient mood

response at Time 2 to expect a Time 3 mood response to have been elicited in
those predisposed, or cognitively vulnerable, to do so.

However, while this interpretation of the results of this study is wholly
consistent with the data, it can not be conclusively derived from the data. In
order to more strongly support the contention that this study confirms a trait
theory for depression and refutes the postulates of HTD, it would be necessary
to conduct a similar study and establish statistically that in a sample divided into
depressed individuals, remitted depressives and normal controls, only those
who are already depressed and who have a depressive attributional style and
who experience a negative life event show a significant enduring increase in
depressive mood. It would be helpful to show that the effect for this group is
strong enough to still show the effect when the depressed group is combined
with the nondepressed groups, because this would explain the results found by
Metalsky et al. (1987). Without this evidence, and in light of other considerations
that will now be discussed, this explanation for the results of this study must be
viewed as one of several interpretations of this study that are consistent with the
data.

A second explanation for the results of this study is that by not
demonstrating a Time 3 MAACL-R depression score effect, the ability to media
such an effect could not be tested. Data consistent with viewing this missing

effect as a simple manipulation failure will now be presented.

92

There are data regarding students’ attained grades on their exams that
may help explain the absence of a Time 3 MAACL-R depression score effect.
As shown in Table 21, no students in either class Go, of all 343 that took an
exam) achieved a failing grade on their exam.24 This suggests that the grading
in these classes may have been viewed by students as being relatively lenient.
Such a view by students would be consistent with the fact that only four
students in the research sample achieved an exam grade lower than their grade
aspiration as reported at baseline (this finding is discussed further on page 95,
below).

Another aspect of the grades received by students may help explain the
absence of a Time 3 MAACL-Fl depression score effect. Of the 84 students in
the unselected sample who received the lowest grades (i.e., a grade of 2.0, 1.5
or 1.0), 68 were selected out of the sample due to missing data, and an
additional 13 were selected out due to high baseline Life Event Inventory (LEI)
and/or BDI scores. This left three students with a grade of 2.0 as the lowest
scoring subjeCts in the research sample. Therefore, those students who would
be expected to be most likely to view their exam outcome as an INLE were
selected out of the study.

Finally, there is data that suggests students’ actual attributions for their
exam outcomes are consistent with their not demonstrating an enduring
depressive mood response to that event. Students’ PAQ scores uniquely
predicted over 28% of the variance in their Time 2 MAACL-R depression scores

for those with a negative exam outcome. The sign of the correlation (pr=-.53),

93

Table 21

 

 

Achieved grade on midterm exam

 

4.0

3.5

3.0

2.5

2.0

1.5

1.0

 

Unselected

sample
(n-343)

32

63

76

61

18

 

Sample selected
for missing
data n-(lOB)

35

11

24

22

13

 

Fully selected
research sample
(n-43)

13

10

12

 

Positive exam
outcome
(n-26)

12

 

 

Negative exam
outcome
(n-l7)

 

 

 

 

 

 

 

 

 

 

94

however, suggests that the relationship between PAQ score and Time 2
MAACL-R depression score was opposite of that which would be viewed as
indicating depressogenic attributions. Table 18 shows that the nature of
student’s PAQ scores was for those with a negative exam outcome to make
relatively lees stable and global attributions, and for those with a positive exam
outcome to make relatively more stable and global attributions. Therefore, bell)
groups’ attributional postures appear to be eebLesemLesjetmj given the
context of their exam outcome (i.e., positive or negative; refer to Figure 3, page
8, regarding depression resistant attributional styles). These findings are
consistent with the absence of an elevation in Time 3 MAACL-R depression
score for both positive and, in particular, negative exam outcome groups.
Taken together, a) the data supporting the possibility that students may
have viewed their exams as having been graded leniently relative to their
expectations, b) the loss of a large proportion of students who achieved a poor
grade from the research sample, and c) the demonstration by students of
relatively depression resistant attributions for their exam performances, can all
be viewed as suggesting that a manipulation failure may have occurred in this
study, thus rendering HTD neither supported nor challenged by the results of
the study. While this interpretation of the results must also be viewed as
consistent with the data, it too can not be conclusively derived from the data.
Again, research of the type suggested earlier (see p. 91) would help to better

clarify how to characterize the results.

95

An important methodological issue regarding how to determine exam
outcome based group assignment is raised by the low number of students who
achieved an exam grade lower than their grade aspiration as reported at
baseline. Metalsky et al. (1987) used a weighted difference score [i.e., (grade
aspiration as reported at baseline - achieved grade) * reported importance of
achievement events on the A80] to indicate the degree to which one found the
receipt of their grade to be an INLE. This study used students’ reports on the
FAQ of being “happy” or “not happy“ with their exam grade as an indicator of
positive vs. negative exam outcome. The two methods correlated r=1.00 in the
Metalsky et al. (1987) study with regard to exam outcome based group
assignment (G. l. Metalsky, personal communication, 1988), but only correlated
r=.10 (ns) in this study. When the PAQ method is applied to the research
sample of this study, 26 students are rated as having experienced a positive
exam outcome and 17 as having experienced a negative exam outcome. When
the weighted difference method is used, 32 subjects are rated as experiencing a
positive exam outcome, four as experiencing a negative exam outcome, and
seven are lost due to failure to complete the grade aspirations questionnaire.
Therefore, the weighted difference method identified fewer students as having
experienced a negative exam outcome than did the PAQ method.

Clearly, the two methods of determining exam outcome based group
assignment did not function as comparably in this study as they did in the
Metalsky et al. (1987) study. While every effort was made to implement both

methods in this study just as they were in the Metalsky et al. (1987) study, it

96

must be considered that one or the other method may have malfunctioned in
some manner in one or both of the studies. Another possible explanation is that
the FAQ measure of exam outcome may have a different criterion for
discriminating positive vs. negative exam outcome than does the weighted
difference method. For example, the weighted difference method may require
the detection of a relatively more conspicuous negative experience than the
PAQ method to rate it as negative, and thus may assign those who have a
mildly negative experince to a positive experience group. Conversely, the FAQ
method may include students who experienced the receipt of their exam grade
as a transitory, relatively unimportant negative life experience (e.g., students
who might say “This bad grade has me feeling down today“) in a group that
was intended to include only those students who experienced the receipt of
their exam grade as an INLE (e.g., students who might say “With this bad
grade, I’ll never be able to succeed“). However, there is insufficient data to
conclusively determine exactly why the PAQ method and the weighted
difference method functioned so differently in the two studies. Therefore, for the
purposes of future research, it would be advisable not to view the two methods
of determining exam outcome based group assignment as being equivalent,
and to use both methods until further research is conducted on the assessment
of INLEs due to negative exam outcomes.

Along similar methodological lines, it is interesting to note that while HTD
(and even LHTD-Fl) clearly stipulate that attributional style and the experience of

an INLE betb interact to predict the hopelessness subtype of depression, little

97

research has focused on the appropriate measurement of lNLEs while
considerable research has been done on measuring attributional style.
Research has been reported on various important issues pertaining to the
measurement of attributional style, for example, that having a depressogenic
attributional style as measured on the ASQ is equally characteristic of non-
depressed social phobics, suggesting that having a depressogenic attributional
style may not be specific to depression (Heimberg, Klosko, Dodge, Shadick,
Becker and Barlow, 1989); that other measures of depressogenic attributional
style with statistically orthogonal scales are more appropriate than the ASQ
because the non-orthogonality of the subscales on the A80 do not reflect the
theoretical orthogonality of the concepts of globality, specificity, and intemality
of HTD (Hill and Kemp-Wheeler, 1986); that depressogenic attributional style
might be more appropriately measured with regard to real events experienced
by subjects rather than with regard to the hypothetical events on the ASQ
(Norman and Antaki, 1988); and that cross cultural differences in how
attributional styles for negative vs. positive events relate to depressive
symptomatology may indicate that depressogenic attributional style might be
best measured as the relative degree to which ones attributional style for
negative events differs from ones attributional style for all types of events
(Crittenden and Lamug, 1988). Yet there is a dearth of research that specifically
addresses the measurement of lNLEs in the context of HT 0. Questions such
as, what are the specific perceived qualities of events that effectively activate the

diathesis proposed by the Diathesis-Stress component of HTD, and at what

98

mamitude must they be experienced, would be appropriate areas of
'mestigation.

Further methodological considerations for the investigation of HTD are
raisml by a recent statement/reconceptualization of the theory by Abramson,
Meldsky and Alloy (1989). Abramson et al. (1989) 'deemphasize causal
attrilutions because inferred negative consequences and inferred negative
characteristics about the self are also postulated to contribute to the formation
of lmelessness and, in turn, the symptoms of hopelessness depression.“ (p.
358)This change in emphasis introduces into the etiological chain proposed by
HTDIhe notion of individuals having depressogenic inferential styles for inferring
negaive consequences of lNLEs and/or for inferring negative characteristics
abut the self given the occurrence of an INLE. Abramson et al. (1989) state
In making either of these types of inferences regarding an INLE can function
sinhrly to making a depressogenic causal attribution about an INLE in the
process of contributing to the formation of hopelessness.

Therefore, just as this study was designed to control for the baseline
concitions of INLEs, depressogenic attributional style, and non-hopelessness
swtypes of depression in order to provide an adequate test of HTD as it was
oriy'lally stated (Abramson et al. 1986, Abramson et al. 1988, Alloy et al.
1%8), studies of HTD as it is now stated by Abramson et al. (1989) will have to
Ms the measurement and control of the baseline conditions for the two
depressogenic inferential styles introduced in that statement of the theory.

Simiarly, even if this study had found a strong Time 3 MAACL-R depression

99

score effect to have occurred, and if all of its hypotheses had been supported,
it would now be appropriate only to view the study as an adequate test of an
earlier version of HTD, and not of the newest version, because it did not
measure and control for the two depressogenic inferential styles introduced by
Abramson et al. (1989).

Abramson et al. (1989) state “We anticipate further expansions and
revisions of the hopelessness theory.“ (p. 365) Given the apparent fluidity of
HTD in the current stage of its development, it is likely that further tests of the
theory as a whole will come to be viewed as obsolete relative to even newer
versions of the theory. While the speculation and research regarding the distal
components of the etiological chain proposed by HTD has had heuristic value in
the effort to understand the cognitive factors that may be antecedent,
concomitant or consequent to possible subtypes of the heterogeneous disorder
of depression, an approach of more enduring value would be to investigate the
theory systematically in terms of its logical components to see if they merit i
continued inclusion in the theory.

For example, as much of the past research on l-ITD and LHTD-R can be
viewed as examinations of some of the more distal components of the
etiological chain proposed by HTD (eg. the relationship between attributional
style and depression), the most proximal component a.e., the relationship
between hopelessness and depression) has been virtually ignored. While
discussing HTD and Beck’s (1967, 1976) cognitive theory of depression, Alloy,

Hartlage and Abramson (1988b) state “Indeed, we know of no work to date that

100

tests whether hopelessness and the negative cognitive triad are, in fact,
sufficient causes of depression or whether they mediate the effects of the more
distal diathesis and stress components of the theories.“ (p. 60). It would seem
logical then, for future research on HT D to investigate this most proximal
component of HT D (after which the theory is named) to see if]; merits further
inclusion in the theory.

In summary, the results of this study were perfectly consistent with the
trait theory of depression. Further, this study did not provide support for HTD.
However, the results were considered in the context of a subject sample that
did not demonstrate any patterns of an enduring depressive mood response.
Two explanations were offered for this unexpectedly absent mood response.
The first explanation suggested that all necessary conditions for testing HTD
were met and that the absence of the enduring mood response was due to
sample selection for recent lNLEs and current depression. This interpretation
was considered to be supportive of the trait theory of depression and of the
views of Barnett and Gotlib (1988), and to hold unfavorable implications for
HTD. The second explanation suggested that even though 40% of the subjects
in the research sample reported being “not happy" with their grade at Time 2
0mmediately after receiving their grade), the absence of a Time 3 enduring
mood response meant that certain hypotheses of this study could not be
tested. This interpretation was presented as neither supportive nor challenging
to HTD, the trait theory of depression, or to the views of Barnett and Gotlib

(1988). Also discussed was how the two methods for determining exam

101
outcome based group assignment (i.e., the PAQ method and the weighted

difference method) functioned quite differently in this study while they functioned
identically in the Metalsky et al. (1987) study. Possible explanations for this
difference were considered. Tne implications for future research involving the
measurement of lNLEs in the context of tests of HTD, and of investigating HTD
in light of its most recent statement Ge, Abramson et al., 1989) were also

presented.

FOOTNOTES

1. Abramson et al. (1978) state that helplessness is the consequence of when
people expect that their future responses will be futile in obtaining a desired
future outcome. However, they make a distinction between pereenel
belojeeseeee, for people who view their situation as helpless yet also believe
that relevant peers would likely be able to perform a response in their repertoire
which would bring about the desired outcome, and bn_iyereel_helple_smees, for
pe0ple who view their situation as helpless and believe that relevant peers
would find that they too do not have a response in their repertoire that would
bring about the desired outcome.

2. Since LHTD, LHTD-R, and the subsequent interpretations of these theories
primarily address the etiological factors of depression, a formal term for this
type of attribution and the tendency to make it has not emerged consistently in
the literature. The term Wm describes this type of attribution in
a fashion that is consistent with the suggestion by Abramson et al. (1978, p.70)
that one therapeutic strategy for treating depression is to help people “change
unrealistic attributions for failure toward external, unstable, specific factors, and
change unrealistic attributions for success toward internal, stable, global
factors.”

3. Abramson et al. (1986), Abramson et al. (1988) and Alon et al. (1988)
distinguish between the occurrence of a negative life event and an individual’s
'attachment“ ofjmbgbtebee to the event after its occurrence. This distinction is
maintained in Figure 5 (a schematic diagram of the etiological chain of
hopelessness depression; see page 14). But for the sake of brevity, for
example, rather that refer to “the experience of a negative life event and the
attachment of importance to that event,’ I will refer to ‘the experience of an
INLE’ 0e, important negative life event).

Also, “INLE“ will be used to refer to the nonoccurrence of an important
positive life event as well as the occurrence of an important negative life event.
This is consistent with the logic of HTD as presented in Abramson et al. (1986),
Abramson et al. (1988) and Alon et al. (1988), and corresponds with their

102

103

similar definition and usage of the terms "negative event,“ “negative life event"
and “important negative life event" in those papers.

4. First, this example assumes that attributing failure to a lack of effort is an
internal, unstable and possibly specific attribution. The purpose of this example
is to suggest that attributing failure to a lack of ability would be an internal,
stable, and possibly global attribution, and that this is more consistent with what
would be expected to lead to a lowered self-esteem.

Second, Alloy et al. (1988) state that in HTD ”when negative life-events
are attributed to internal, as well as stable, global causes, hopelessness will be
accompanied by lowered self-esteem (Crocker, Alloy & Kayne, 1987; Tennen &
Herzberger, 1987)," (p. 9) but they do not explicitly present this as different from
LHTDR. Similarly, Abramson et al. (1988) state that “when negative life events
are attributed to internal as well as stable, global causes, Abremeen et el.
[(1918)] hypothesized that the expectation of hopelessness will be accompanied
by lowered self-esteem." (p. 10, emphasis added) In Fact, Abramson et al.
(1978) state that "whether self-esteem is lowered will depend on the intemality
of the attribution for helplessness“ (p. 68), with no direct regard given to the
globality or stability of the attribution. The statements of Alon et al. (1988) and
Abramson et al. (1988) are vague in that they may be read as consistent with
Abramson et al. (1978) (e.g., the cognitive, motivational and affective
hopelessness deficits, which depend on the stability and globality of attributions
for uncontrollable events, will be accompanied by lower self-esteem, which
depends only on the intemality of attributions, if the uncontrollable events are
attributed to internal, global, stable causes). However, if this is the case, these
statements are inconsistent with Abramson et al. (1986). I have chosen to refer
to the Abramson et al. (1986) interpretation of this issue in the text because it is
the least vague and it clearly addresses its position as a revision of LHTD-R.

5. In Figure 1, HT D allows for the possibility that hopelessness might, in some
instances, be elicited by causes other than the proposed chain of contributory
causes. So the chain of contributory causes may be viewed as one of several
possible causes of hopelessness, just as hopelessness is viewed as one of
several possible causes of depression.

6. Since HT D states that the minimel conditions for viewing a causal attribution
for an INLE as depressogenic is for that attribution to be small and eteble in
nature, it is important to define how references to depressogenic attributions
and attributional styles are defined for this paper. The term “depressogenic
attribution" will refer to the actual attribution of an INLE to a global and stable
cause. The term ”depressogenic attributional style” will refer to the tendency to
make stable and global depressogenic attributions for lNLEs. When the
intemality of a depressogenic attribution or depressogenic attributional style is

104

of issue, this will be stated clearly [e.g., by refering to a “depressogenic
attribution (internal, as well as global and stable)’].

7. The practice of using the principles of HTD to critique investigations of
LHTD-R is quite defendable. Besides the former being in large part a
clarification of the later, it has been noted that typical research on LHTD-R
actually conforms better to the proposed principles of HTD. For example, Alloy
et al. (1988, p. 8) note that "the majority of [studies investigating LHTD-R]
focused on the occurrence of negative life-events rather than uncontrollable
events...“ In this regard, such studies could be more reasonably viewed as
inadvertent tests of HTD than as tests of LHTD-R.

8. The "WW“ for hopelessness depression that

Abramson et al. (1986; p. 49) refers to is a legacy of the cognitive, motivational
and emotional deficits that Seligman (1975) observed in dogs who had been
exposed to inescapable electric shock. Seligman (1975) founded LHTD on his
observance that these deficits were eimiler to those demonstrated by depressed
humans. To the degree that Seligman’s observations were accurate, it is
unlikely that the symptoms displayed by people experiencing hopelessness
depression (as defined by the process which produces it) will be extraordinarily
different from those displayed by people who are experiencing a non-
hopelessness subtype of depression. However, efforts to determine the nature
of the depressive symptoms associated with hopelessness depression could be
termed tests of a third component of HTD, namely the W

W. However, tests of this symmem component would have to
be done after tests of the 2:09.855 components were completed.

9. HTD had not been formally published at the time the Metalsky et al. (1982)
study was conducted and published. However, since Gerald l. Metalsky and
Lyn Y. Abramson are among the primary HT D theorists (with Lauren 8. Alloy), it
is likely that the Metalsky et al. (1982) study was conducted in light of an
informal though advanced conceptualization of HTD. In a critique of that study,
Williams (1985) commended the authors for their unambiguous statement of
LHTD-R, a statement that is clearly a precursor of HTD.

10. The original ASQ was presented by Peterson, Semmel, von Baeyer,
Abramson, Metalsky and Seligman (1982), and had three items for each of the
following goal outcomes: good achievement, bad achievement, good affiliation
and bad affiliation. The ASQ used by Metalsky et al. (1987) had six items each
for bad achievement and bad interpersonal outcomes. This was done to
increase the reliability of the two bad outcome subscales so they could be
examined separately. Metalsky et al. (1987) did not use the good outcome
subscales in their version of the ASQ.

105

11. In this quote, Metalsky et al. (1987, p.392) cite “Weiner (1986)“ and
reference Weiner’s work as a manuscript submitted for publication. In fact,
Weiner’s work was published in W in 1985, and is
referenced as such in this paper.

12. Table 5 is sufficient to illustrate this point. It demonstrates all possible
relationships between whether one is depressed, has a depressogenic
attributional style or has experienced an INLE. It is not necessary to breakdown
each example further by splitting them based on whether one has a non-
hopelessness subtype of depression. This is because hopelessness depression
(and correspondingly non-hopelessness depression) is defined by its etiological
process and therefore cannot be identified without information about both
attributional style and INLE's. Hence, the situation of having information about
the baseline incidence of non-hopelessness depression in the absence of
information about the baseline incidence of attributional style or INLE’s is
empirically indeterminable (until a reliable covariate for one of these variables is
established).

13. Two subjects of the 43 students in the research sample did not complete
the measures of hopelessness. Therefore, neither of them are represented in
the tests of Causal Mediation H‘ or H5. Since these tests use only a portion of
the research sample and only one of these subjects were in that portion of the
sample, their absence is reflected by a change of only one degree of freedom
in those tests.

14. Metalsky et al. (1987) measured student’s exam outcome by calculating the
degree to which students’ actual exam grades exceeded or fell beneath the
grades students had reported they would consider a failure (G. I. Metalsky,
personal communication, 1988). This was viewed as a measure of students”
subjective evaluation of their success or failure on the exam. They then
multiplied this value by students’ scores on the importance subscale of the ASQ
to weight the exam outcome measure more heavily for those who found such
achievement oriented events to be more important to them. This method
appeared valid, especially since there was 3 +1.00 correlation between the
classification of students into good vs. bad exam outcome groups using this
method and the separation of students into groups of “happy“ vs. ”not happy"
with their grade based on the reports by students on the PAQ within ten
minutes of receiving their grades (G. l. Metalsky, personal communication,
1988). This study had intended to use the same method as Metalsky et al.
(1987) to measure students’ exam outcomes, but for this study that method
correlated with the PAQ method only .10 (p=.55) with regard to exam outcome
based group assignment. Therefore, the author chose to use the PAQ method
since it is an immediate self-report of students’ actual subjective evaluations of
their exam outcomes. A grade aspiration questionnaire (Metalsky et al., 1987;

w!— mm may maul”

106

see Appendix C) was administered to subjects during an assessment session in
class early in the baseline period of this study with the intent of duplicating the
exam outcome measure used by Metalsky et al. (1987). However, since the
PAQ method was used, the grade aspiration questionnaire served no utility and
will not be further addressed.

15. Differences in assessment methods between this study and that of Metalsky
et al. (e.g., using the MAACL-R rather than the MAACL) have already been
noted as each measure was discussed in the method. Notable procedural
differences between this study and that of Metalsky et al. (1987) were as
follows. Subjects in this study were enrolled in one of two different classes and
therefore took one of two different exams, as opposed to one class and one
exam for the Metalsky et al. (1987) study. This study averaged two or more pre-
exam MAACL-R scores to calculate baseline levels of depression while the
Metalsky et al. (1978) study used a single implementation of the MAACL given
two days ef_te_r students took their exam but three days before they received
their exam grades (they note that similar results were obtained using a baseline
MAACL given two days before students took their exam). This study included
one implementation each of the ASQ, BDI, LEI and BHS in an out of class
testing session (and another BHS in class at the end of the study) while the
Metalsky et al. (1987) implemented only the ASQ out of class and did not use
the other measures. This study took place during mid-Spring of 1988 using 43
students at Michigan State University while the Metalsky et al. (1978) study took
place during late-Winter of 1982 using 94 students at the University of
Wisconsin.

16. In both classes after all data were collected, students were asked to
hypothesize what the nature was of each of the two apparent experiments. In
both classes, no one reported speculating that the two experiments were
related.

17. The PAQ measures the degree to which an attribution for exam
performance is global and stable and does not measure how depressogenic the
attribution is. A global and stable attribution for a negative outcome is viewed to
be depressogenic and a global and stable attribution for a positive outcome is
viewed to be depression resistant.

18. The n’s are unequal for the baseline implementations of the MAACL-R
because not every student attended every implementation. During the baseline
period, the MAACL-R was given seven times in the introductory psychology
class and five times in the psychology statistics class. The first assessment in
each class was dropped to reduce any novelty effect. Student’s baseline
depressive moods were estimated by taking the mean of remaining baseline
measures with a minimum of two values required. The first two MAACL-R

107

depression score reliabilities reported have n’s of 22 and 23 because they apply
only to the students in the introductory psychology class, of which 23 were in
the research sample. The other reliabilities applied to assessments in both
classes so their n’s of 41, 40, 34 and 40 are relative to the total research
sample of 43 students.

19. The homogeneity of regression test (see Appendix K) for the analysis of
partial variance testing the Time 2 comparison to Causal Mediation H did
approach significance. In this test, the set of factors testing homogeneity of
regression correlated pr=.62 (p=.07) with the dependent variable. However,
since this analysis of partial variance was only a comparison to a hypothesis
test for this study, no further analysis was pursued.

20. This result is called into question by the near failure of the test for
homogeneity of regression for this analysis of partial variance (see footnote 19).

21. This hypothesis, as originally stated, neglected to specify that this analysis is
most appropriately done with specific regard to those students with a negative
exam outcome. Such an analysis is logically consistent with all other analyses in
this study that used immediate or enduring MAACL-R depression scores as a
dependent variable and which did not use exam outcome as an independent
variable. Therefore, to facilitate both the logical and interpretive consistency of
this study, this analysis was done using only data from students with a negative
exam outcome. ’

22. For identical reasons as described in footnote 19 regarding Causal
Mediation H‘, this analysis was done using only data from students with a
negative exam outcome.

23. The Metalsky et al. (1987) study used the original MAACL-Today Form in
their study while this study used the revised version. Both versions have the
same 132 items on the check list (except that item 52 was changed from ”gay"
to 'Iively"), but the revised version 0.e., the MAACL-R) has a more current and
thorough restandardization.

24. Exam grades in the two classes were given on a four point basis, as is
customary at Michigan State University (i.e., 0.0=failing to 4.0=A: 0.0 and 1.0 to
4.0 in .5 increments).

APPENDIX A

Al’l I I I Inlr

ASQ

STUDENT ID! Date

 

Mien;

Please try to vividly imagine yourself in each of the situations
or sequences of events that follow. Picture each situation as clearly
as you can and as if the events were happening to you right now. Place
yourself in each situation and decide what you feel would have eebeee
it if it actually happened to you. Although events may have many
causes, we want you to choose only bne -- the meie; cause if the event
actually happened to you. For each situation, you will write down this
cause in the blank provided. Then we will ask you some questions about
the cause.

It is important to remember that there are no right or wrong
answers to the questions. The important thing is to answer the
questions in a way that corresponds to what yeg would think and feel
if the situations actually were occurring in your life.

108

109
Page 2

A” Imagine that the following sequence of events eebgelly happens to
you:

 

You take an exam and receive a low grade on it.
The questions below ask about the eeeee of your low grade on the exam.

First, write down the ene major cause of your low grade on the
exam.

 

1. Is it something about you or something about other people or
circumstances that caused your low grade on the exam? (Choose one
number.) . .

Totally caused Totally caused
by other people 1 2 3 4 S 6 7 by me
or circumstances

2. In the future when taking exams, will the cause of the low
grade on this exam also cause other exam grades of yours to be
low? (Choose one number.)

 

Will never again Will always
cause my exam 1 2 3 4 5 6 7 cause my exam
grades to be low grades to be low

3. Is the cause of your low grade on the exam something that
causes problems just in that exam grade, or does it also cause
problems in other areas of your life? (Choose one number.)

Causes problems Causes problems
just in that l 2 3 4 5 6 7 in all areas
exam grade of my life

4. How important is it to you that your grade on the exam is low?
(Choose one number.)

Not at all 1 2 3 4 5 6 7 Extremely
important important

110
Page 3

8. Imagine that the following sequence of events 5222311! happens to
you:

You don't have a boyfriend/girlfriend (or spouse) although you want
one.

The questions below ask about the eegee of your not having a
boyfriend/girlfriend (or spouse) although you want one.

First, write down the egg major cause of your not having a
boyfriend/girlfriend (or spouse) although you want one.

 

5. Is it something about you or something about other people or
circumstances that caused your not having a boyfriend/girlfriend
(or spouse) although you want one? (Choose one number.)

Totally caused Totally caused
by other people 1 2 3 4 5 6 7 by me
or circumstances

6. In the future when you want a boyfriend/girlfriend (or
spouse), will the cause of your not having a boyfriend/girlfriend
(or spouse) now also cause you not to have a boyfriend/girlfriend
(or spouse) then? (Choose one number.)

Will never again Will always
cause me to not 1 2 3 4 S 6 7 cause me to not
have a boyfriend/ have a boyfriend/
girlfriend (or girlfriend (or
spouse) spouse)

7. Is the cause of your not_having a boyfriend/girlfriend (or
spouse) something that causes problems just in that instance of
wanting a boyfriend/girlfriend (or spouse), or does it also cause
problems in other areas of your life? (Choose one number.)

Causes problems Causes problems
just in that l 2 3 4 5 6 7 in all areas
instance of wanting of my life

a boyfriend/girlfriend
(or spouse)

8. How important is it to you that you don't have a
boyfriend/girlfriend (or spouse) although you want one (Choose
one number.)

Not at all 1 2 3 4 5 6 7 Extremely
important important

111
Page 4

C. Imagine that the following sequence of events eebgelly happens to
you:

A friend comes to you with a problem, and you are not as helpful as
you would like to be.

The questions below ask about the beeee of your not being as helpful
as you would like to be to your friend.

First, write down the 922 major cause of your not being as
helpful as you would like to be to your friend.

 

9. Is it something about you or something about other people or
circumstances that caused your not being as helpful as you would
like to be to your friend? (Choose one number.)

Totally caused Totally caused
by other people 1 2 3 4 5 6 7 by me
or circumstances .

10. In the future when a friend comes to you with a problem, will
the cause of your not being as helpful as you would like to be to
your friend now also cause you to not be as helpful as you would
like to be to a friend then? (Choose one number.)

Will never again Will always cause
cause me to not 1 2 3 4 5 6 7 me to not be as

be as helpful as helpful as I would
I would like to be like to be

11. Is the cause of your not being as helpful as you would like
to be to your friend something that causes problems just your
helping that friend, or does it also cause problems in other
areas of your life? (Choose one number.)

Causes problems Causes problems
just in my 1 2 3 4 5 6 7 in all areas
helping that of my life
friend

12. How important is it to you that you are not as helpful as you
would like to be to your friend? (Choose one number.)

Not at all 1 2 3 4 5 6 7 Extremely
important important

112
Page 5

D. Imagine that the following sequence of events eebgelly happens to
you: '

As an assignment, you give an important talk in class, and the class
reacts negatively. ,

The questions below ask about the eeeee of the class reacting
negatively to your talk.

First write down the bee major cause of the class reacting
negatively to your talk.

 

13. Is it something about you or something about other people or
circumstances that caused the class to react negatively to your
talk? (Choose one number.)

Totally caused . Totally caused
by other people 1 2 3 4 5 6 7 by me
or circumstances

14. In the future when you give important talks in class, will
the cause of the class reacting negatively to this talk also
cause the class to react negatively to other talks of yours?
(Choose one number.)

Will never again Will always

cause the class 1 2 3 4 5 6 7 cause the class to
to react negatively react negatively
to my talks to my talks

15. Is the cause of the class reacting negatively to your talk
something.that causes problems just in that instance of giving a
talk, or does it also cause problems in other areas of your life?
(Choose one number.)

Causes problems Causes problems
just in that 1 2 3 4 5 6 7 in all areas
instance of of my life

giving a talk

16. How important is it to you that the class reacts negatively
to your talk? (Choose one number.)

Not at all 1 2 3 4 5 6 7 Extremely
important important

 

 

L.

113
Page 6

E. Imagine that the following sequence of events 39533112 happens to
you:

 

Your parents have been treating you in a negative way.

The questions below ask about the eeeee of your parents treating you
in a negative way.

First write down the ene major cause of your parents treating you
in a negative way.

 

17. Is it something about you or something about other people or
circumstances that caused your your parents to treat you in a
negative way? (Choose one number.)

Totally caused Totally caused E?
by other people 1 2 3 4 5 6 7 by me in“.
or circumstances ' '

18. In the future when interacting with your parents, will the
cause of them treating you in a negative way also cause them to

 

treat you in a negative way then? (Choose one number.) [yin
Will never again Will always‘ '
cause my parents 1 2 3 4 S 6 . 7 cause my parents

to treat me in a to treat me in a

negative way negative way

19. Is the cause of your parents treating you in a negative way
something that causes problems just in that instance of
interacting with them, or does it also cause problems in other
areas of your life? (Choose one number.)

Causes problems Causes problems
just in that l 2 3 4 5 6 7 in all areas
instance of of my life
interacting with

my parents

20. How important is it to you that your parents have been
treating you in a negative way? (Choose one number.)

Not at all 1 2 3 4 5 6 7 Extremely
important important

114
Page 7

F. Imagine that the following sequence of events 39523112 happens to
you:

Your gradepoint average (GPA) for the semester is low.

The questions below ask about the eeeee of your low gradepoint average
(GPA) for the semester.

First write down the bee major cause of your low gradepoint
average (GPA) for the semester.

 

21. Is it something about you or something about other people or
circumstances that caused your low gradepoint average (GPA) for
the semester? (Choose one number.)

Totally caused Totally caused
by other people 1 2 3 4 5 6 7 by me
or circumstances

22. In the future when you receive your grades for a semester,
will the cause of this semester's low gradepoint average (GPA)
also cause other semesters gradepoint averages (GPA's) of yours
to be low? (Choose one number.)

Will never again Will always

cause my semester 1 2 3 4 5 6 7 cause my semester
gradepoint averages gradepoint averages
(GPA's) to be low (GPA's) to be low

6

23. Is the cause of your low gradepoint average (GPA) for the
semester something that causes problems just in your gradepoint
average for that semester, or does it also cause problems in
other areas of your life? (Choose one number.)

Causes problems Causes problems
just in my 1 2 3 4 5 6 7 in all areas
gradepoint average of my life

for that semester

24. Bow important is it to you that your low gradepoint average
(GPA) for the semester is low? (Choose one number.)

Not at all 1 2 3 4 5 6 7 Extremely
important important

115
Page 8

G. Imagine that the following sequence of events eebeelly happens to
you: .

At a party, people don't act interested in you.

The questions below ask about the eebee of people not acting
interested in you at the party.

First write down the 929 major cause of people not acting
interested in you at the party.

 

25. Is it something about you or something about other people or
circumstances that caused people to not act interested in you at
the party? (Choose one number.)

Totally caused Totally caused
by other people 1 2 3 4 5 6 7 by me
or circumstances

26. In the future when at parties, will the cause of peOple not
acting interested in you at this party also cause people to not
act interested in you at other parties? (Choose one number.)

Will never again Will always

cause people to 1 2 3 4 5 6 7 cause people to to
not act interested not act interested
in me at parties in me at parties

27. Is the cause of people to not act interested in you at the
party something that causes problems just in peOple's interest in
you at that party, or does it also cause problems in other areas
of your life? (Choose one number.)

Causes problems Causes problems
just in people's 1 2 3 4 5 6 7 in all areas
interest in me of my life

at that party

28. How important is it to you that at a party, people don't act
interested in you? (Choose one number.)

Not at all 1 2 3 4 5 6 7 Extremely
important important

 

 

116
Page 9

H. Imagine that the following sequence of events 55533111 happens to
you: ,

You write a paper for a course and get a low grade on it.

The questions below ask about the 25229 of your getting a low grade on
your paper.

First write down the bee major cause of your getting a low grade
on your paper

 

29. Is it something about you or something about other people or
circumstances that caused your getting a low grade on your paper?
(Choose one number.)

Totally caused Totally caused
by other people 1 2 3 4 5 6 7 by me
or circumstances

30. In the future when writing papers for a course, will the
cause of your getting a low grade now also cause you to get a low
grade then? (Choose one number.) .

Will never again Will always
cause me to get 1 2 3 4 5 6 7 cause me to get
a low grade a low grade

31. Is the cause of your getting a low grade on your paper
something that causes problems just in that instance of writing
papers, or does it also cause problems in other areas of your
life? (Choose one number.)

Causes problems Causes problems
just in that 1 2 3 4 5 6 7 in all areas
instance of of my life

writing papers
for a course

32. How important is it to you that you got a low grade on a
paper for a course? (Choose one number.)

Not at all 1 2 3 4 5 6 7 Extremely
important important

117
Page 10

1. Imagine that the following sequence of events eebeelly happens to
you:

Your professor gives an important lecture, and you don't understand
it.

The questions below ask about the eegee of your not understanding the
lecture.

First write down the bee major cause of your not understanding
the important lecture.

 

33. Is it something about you or something about other people or
circumstances that caused you to not understand the important
lecture. (Choose one number.)

Totally caused Totally caused
by other people 1 2 3 4 5 6 7 by me
or circumstances

34. In the future when listening to important lectures, will the
cause of your not understanding the lecture now also cause you to
not understand important lectures then? (Choose one number.)

Will never again Will always

cause me to not 1 2 3 4 S 6 7 cause me to not
understand an understand
important lecture important lectures

35. Is the cause of your not understanding the important lecture
something that causes problems just in your understanding that
important lecture, or does it also cause problems in other areas
of your life? (Choose one number.)

Causes problems Causes problems
just in my 1 2 3 4 5 6 7 in all areas
understanding that of my life

important lecture

36. How important is it to you that you don't understand that
important lecture? (Choose one number.)

Not at all 1 2 3 4 5 6 7 Extremely
important important

118
Page 11

J. Imagine that the following sequence of events eebeelly happens to
you:

Your teacher asks a question in class, and you don't know the answer.
The questions below ask about the eeeee of your not knowing the answer
to your teacher's question.

First write down the 982 major cause of your not knowing the
answer to your teacher's question?

 

37. Is it something about you or something about other people or
circumstances that caused you to not know the answer to your
teacher's question? (Choose one number.)

Totally caused Totally caused
by other people 1 2 3 4 5 6 7 by me
or circumstances

 

38. In the future when your teacher asks a question in class,
will the cause of your not knowing the answer to this question
also cause you to not know the answer to the other questions?
(Choose one number.) H

 

Will never again Will always
cause me to not 1 2 3 4 5 6 7 cause me to not
know the answer know the answer

39. Is the cause of your not knowing the answer to your teacher's
question something that causes problems just answering that
question, or does it also cause problems in other areas of your
life? (Choose one number.)

Causes problems Causes problems
just in answering l 2 3 4 5 6 7 in all areas
that question of my life

40. how important is it to you that when your teacher asks a
question in class you do not know the answer? (Choose one

number.)

Not at all 1 2 3 4 S 6 7 Extremely
important important

119
Page 12

K. Imagine that the following sequence of events eegeelly happens to
you:

Your relationship with your boyfriend/girlfriend (or spouse) ends
even though you would like it to continue.

The questions below ask about the eeeee of your relationship with your
boyfriend/girlfriend (or spouse) ending even though you would like it
to continue.
First write down the bee major cause of your relationship with
your boyfriend/girlfriend (or spouse) ending even though you
would like it to continue.

 

41. Is it something about you or something about other people or
circumstances that caused your relationship with your
boyfriend/girlfriend (or spouse) to end even though you would
like it to continue? (Choose one number.)

Totally caused Totally caused
by other people 1 2 3 4 5 6 7 by me
or circumstances

42. In the future when you are involved in a relationship, will
the cause of your relationship with your boyfriend/girlfriend (or
spouse) ending now also cause other relationships with
boyfriends/girlfriends (or spouses) to end even though you would
like them to continue? (Choose one number.)

Will never again ‘ Will always cause
cause my 1 2 3 4 5 6 7 my relationships
relationships with with boyfriends]
boyfriends/girlfriends girlfriends

(or spouses) to end (or spouses) to end

43. Is the cause of your relationship with your
boyfriend/girlfriend (or spouse) ending even though you would
like it to continue something that causes problems just in your
relationship with your boyfriend/girlfriend (or spouse), or does
it also cause problems in other areas of your life? (Choose one

number.)

Causes problems Causes problems
just in my 1 2 3. 4 5 6 7 in all areas
relationship of my life

with my boyfriend/
girlfriend (or spouse)

44. How important is it to you that your relationship with your
boyfriend/girlfriend (or spouse) ends even though you would like
it to continue? (Choose one number.)

Not at all 1 2 3 4 5 6 7 Extremely
important important

120
Page 13

L. Imagine that the following sequence of events eebeelly happens to
you:

A person with whom you really want to be friends does not want to be
friends with you.

The questions below ask about the eegee of the person not wanting to
be friends with you.

First write down the ene major cause of the person not wanting to
be friends with you.

 

45. Is it something about you or something about other people or
circumstances that caused the person to not want to be friends
with you. (Choose one number.)

Totally caused Totally caused
by other people 1 2 3 4 S 6 7 by me
or circumstances

46. In the future when you want to be friends with someone, will
the cause of this person not wanting to be friends with you also
cause other people to not want to be friends with you? (Choose
one number.)

Will never again Will always

cause other 1 2 3 4 5 6 7 cause other people
people to not want to not want to be
to be friends with me friends with me

47. Is the cause of the person not wanting to be friends with you
something that causes problems just in that person wanting to be
friends with you, or does it also cause problems in other areas
of your life? (Choose one number.)

Causes problems Causes problems
just in that l 2 3 4 5 6 7 in all areas
person wanting to of my life

be friends with me

48. How important is it to you that a person with whom you really
want to be friends does not want to be friends with you? (Choose
one number.)

Not at all 1 2 3 4 S 6 7 Extremely
important important

APPENDIX B

Prtil l’ n I

PAQ

STUDENT 10’: DATE:

 

If you are beppy with your midterm grade, please fill out Questions
1-12 (pages 1-2).

If you are ge§_beppy with your grade, please fill out Questions 13-24
(pages 3-4).

First, what is the egg major cause of you obtaining a high grade on
the Psychology exam? .

 

Is it something about you or something about other people or
circumstances that caused your high grade on the Psychology
exam? (Circle one number)

Totally caused Totally caused
by other people 1 2 3 4 S 6 7 by me
or circumstances

In the future, will the cause of your high grade on this exam
also cause your grade on the next Psychology exam in this
class to be high? (Circle one number.)

Will not cause Will cause my grade
my grade on the 1 2 3 4 5 6 7 on the next

next Psychology psychology exam
exam to be high to be high

In the future, will the cause of your high grade on the exam also
cause your grades on other exams to be high? (Circle one number.)

Will never cause Will always cause
my grades on 1 2 3 4 S 6 7 my grades on other
other exams to exams to be high
be high

Is the cause of your high grade on this exam something that caused
a positive outcome just in this exam grade, or does it also cause
positive outcomes in your grades on exams in other Psychology
courses? (Circle one number)

Caused a positive Causes positive
outcome just in my 1 2 3 4 5 6 7 outcomes in my
grade on this grades on all
Psychology exam exams in other

Psychology courses

121

122
Page 2

5. Is the cause of your high grade on this exam something that caused
a positive outcome just in this exam grade, or does it also cause
positive outcomes in your grades on exams in courses outside of
Psychology? (Circle one number)

Caused a positive Causes positive

_outcome just in l 2 3 4 5 6 7 outcomes in my grades

my grade on this on all exams in

Psychology exam courses outside of
Psychology

6. Is the cause of your high grade on this exam something that caused
a positive outcome just in this exam grade, or does it also cause
positive outcomes in other areas of your life? (Circle one number)

Caused a positive Causes positive
outcome just in 1 2 3 4 5 6 7 outcomes in all
my grade on this situations in my life

Psychology exam

7. Potentially, there are a number of factors that may contribute to
a person receiving a grade with which he/she is happy on this
Psychology exam. Examples of some of these factors are the person's
ability, how hard he/she tried, how easy the exam was, and how
lucky he/she was. Questions 8a-d ask how important each of these
factors was in determining gee; grade on this Psychology exam.

8. How important do you think that ability on your part was in
determining your grade on the Psychology exam? (Circle one number)

Not at all important 1 2 3 4 5 6 7 Extremely important

9. How important do you think that effort on your part or trying hard
was in determining your grade on the Psychology exam? (Circle one
number)

Not at all important 1 2 3 4 5 6 7 Extremely important

10. How important do you think that easiness of the questions on the
Psychology exam was in determining your grade on the Psychology
exam? (Circle one number)

Not at all important 1 2 3 4 5 6 7 Extremely important

11. How important do you think that good luck was in determining your
grade on the Psychology exam? (Circle one number)

Not at all important 1 2 3 4 5 6 7 Extremely important

12. How important is it to you that your grade on the Psychology exam
is high? (Circle one number)

Not at all important 1 2 3 4 5 6 7 Extremely important

123
Page 3

Questions 13-24 are for people who are beb_beepy with their grade.

First, what is the 922 major cause of you obtaining a low grade on the
Psychology exam?

 

13. Is it something about you or something about other people or
circumstances that caused your low grade on the Psychology exam?
(Circle one number)

Totally caused Totally caused
by other people 1 2 3 4 5 6 7 by me
or circumstances 4

14. In the future, will the cause of your low grade on this exam also
cause your grade on the next Psychology exam in this class to be
low? (Circle one number.)

Will not cause Will cause my grade
my grade on the 1 2 3 4 5 6 7 on the next

next Psychology psychology exam
exam to be low to be low

15. In the future, will the cause of your low grade on the exam also
cause your grades on other exams to be low? (Circle one number.)

Will never cause Will always cause
my grades on 1 2 3 4 5 6 7 my grades on other
other exams to exams to be low
be low

16. Is the cause of your low grade on this exam something that caused
problems just in this exam grade, or does it also cause problems
in your grades on exams in other Psychology courses? (Circle one

number)

Caused problems Causes problems
just in my 1 2 3 4 5 6 7 in my grades on
grade on this all exams in
Psychology exam other Psychology

30113898

17. Is the cause of your low grade on this exam something that caused
problems just in this exam grade, or does it also cause problems
in your grades on exams in courses outside of Psychology? (Circle

one number)

Caused a problem ' Causes problems
just in my grade 1 2 3 4 5 6 7 in my grades

on this on all exams in
Psychology exam courses outside of

Psychology

124
Page 4

18. Is the cause of your low grade on this exam something that caused
a problem just in this exam grade, or does it also cause problems
in other areas of your life? (Circle one number)

Caused a problem Causes problems
just in my grade 1 2 3 4 5 6 7 in all situations
on this in my life

Psychology exam

19. Potentially, there are a number of factors that may contribute to
a person receiving a grade with which he/she is ebb happy on this
Psychology exam. Examples of some of these factors are the
person's lack of ability, how little he/she tried, how difficult
the exam was, and how unlucky he/she was. Questions 17a-d ask how
important each of these factors was in determining gee; grade on
.this Psychology exam.

20. How important do you think that a lack of ability on your part
was in determining your grade on the Psychology exam? (Circle one
number)

Not at all important 1 2 3 4 5 6 7 Extremely important

21. How important do you think that a lack of effort on your part or
not trying hard was in determining your grade on the Psychology
exam? (Circle one number)

Not at all important 1 2 3 4 5 6 7 Extremely important

22. How important do you think that difficulty of the questions on
the Psychology exam was in determining your grade on the exam?
(Circle one number)

Not at all important 1 2 3 4 5 6 7 Extremely important

23. How important do you think that bad luck was in determining your
grade on the Psychology exam? (Circle one number)

Not at all important 1 2 3 4 5 6 7 Extremely important

24. How important is it to you that your grade on the Psychology exam
is low? (Circle one number)

Not at all important 1 2 3 4 5 6 7 Extremely important

APPENDIX C

r lrln innl

A9
STUDENT 10’: DATE:

 

lemmas:

Please answer each of the questions that follow. Read each question
carefully before responding.

1. What grade do you expect to receive on the first midterm exam in
this Psychology course? (Choose one answer, a-h)

a)4.0 b)3.5 c)3.0 a)2.s e)2.0 1,1.5 g)l.0 h)0.0

2. What grade would you consider a feileze on the first midterm exam
in this Psychology course? (Choose one answer, a-h)

a)4.0 b)3.5 c)3.0 d)2.5 e)2.0 £)1.s g)1.0 mo.o

3. What grade would you consider a eeeeeee on the first midterm exam
in this Psychology course? (Choose one answer, a-h)

a)4.0 b)3.5 C)3.0 d)2.5 e)2.0 f)1.5 g)1.0 h)0.0

4. What grade would you be egbgppy with on the first midterm exam in
this Psychology course? (Choose one answer, a-h)

”4.0 b)3.5 c)3.0 d)2.5 e)2.0 f)1.5 g)1.0 h)0.0

5. What grade would you be beppy with on the first midterm exam in
this Psychology course? (Choose one answer, a-h)

a)4.0 b)3.5 c)3.0 d)2.5 e)2.0 f)1.5 g)l.0 mo.o

6. What grade on the first midterm exam in this Psychology course
would make you feel : : 0! . a-

W? iChooae one answer. a-il
a)4.0 b)3.5 c)3.0 d)2.5 e)2.0 f)1.5 g)1.0 mo.o

  

i) There is no grade on the first midterm exam that would make
me feel hopeless about getting a good grade on the next midterm
exam in this course.

7. What grade on the first midterm exam in this Psychology course
would make you feel 1"3,- eo-

W? (Coo-cone answer. a-il _—
a)4.0 b)3.5 c)3.0 d)2.5 e)2.0 “1.5 g)1.0 h)0.0

  

i) There is no grade on the first midterm exam that would make
me feel hapeful or confident about getting a good grade on the
next midterm exam in this course.

125

126
Page 2

8. What grade on the first midterm exam in this Psychology course

‘would make you feel W
W? (Choose one answer. e-ll

“4.0 b)3.5 c)3.0 d)2.5 e)2.0 31.5 g)1.0 h)0.0

i) There is no grade on the first midterm exam that would make
me feel hopeless about getting a good grade in the course as a
whole. .

9. What grade on the first midterm exam in this Psychology course

would make you feel W
W? (choose one answer. a-i)

a)4.0 b)3.5 c)3.0 d)2.5 e)2.0 £)1.5 g)l.0 h)0.0

i) There is no grade on the first midterm exam that would make
me feel hopeful or confident about getting a good grade in the
next the course as a whole.

10. What grade on the first midterm exam in this Psychology course
would make you feel WW? (Choose one

answer, a-i)

a)4.0 b)3.5 c)3.0 d)2.5 e)2.0 f)1.5 g)1.0 h)0.0

i) There is no grade on the first midterm exam that would
make me feel hopeless about school in general.

11. What grade on the first midterm exam in this Psychology course
would make you feel W?

(Choose one answer, a-i)
a)4.0 b)3.5 c)3.0 d)2.5 e)2.0 f)1.5 g)1.0 h)0.0

i) There is no grade on the first midterm exam that would
make me feel hopeful or confident about school in general.

 

 

APPENDIX D

MItIIAfl A IV hk -VI T Frm

MULTIPLE AFFECT
ADJECTIVE CHECK LIST

STATE / TODAY FORM

By Mervin Zuckerman
and
Bernard Lubin

Name ...................................... Age ........ Sex ......

Date ....................... Highest grade completed in school ......

DIRECTIONS: On this sheet you will find words which describe different
kinds 0i moods and feelings. Mark an E in the boxes beside the words
which describe how you ieel now - today. Some OI the words may sound

alike, but we want you to check all the words that describe your feelings.

 

Work rapidly.

PUBLISHED av EdITS
'0. am 723.4
SAN DIEGO. CA 92107

Copyright 1.’ 1965 by EleSEducanonal and Industrial Testing Semice. All nghu reserved,
Reproduction of this form by any means ""6th oronibitso.

DIINYIO ll 0.0.A.

6'3 use 001

127

1 C] active

2 E] adventurous
3 D affectionate
4 D afraid

5 [jagitated

6 Dagreeable
7 Daggressive
8 Dalive

9 E] alone

10 [j amiable

ll Damused

12 C] angry

13 Dannoyed

14 [jawful

15 Dbashful

16 Dbitter

17 U blue

18 D bored

19 C] calm

20 C] cautious

21 C] cheerful

22 [3 clean

23 C] complaining
24 [:Jcontented
25 Dcontrary
26 Dcool

27 C] cooperative
28 Ucritical

29 [:1 cross
30 Dcruel
31 Ddaring
32 [jdesperate
33 Ddestroyed
34 Udevoted
35 Cldisagreeable
36 C discontented
37 Ediscouraged
38 "_— disgusted
39 Edispleased
40 Eenergetic
41 : enraged
42 [enthusiastic
43 :fearful

44 : fine

128

PA
45 [I fit

46 Diorlorn
47 Dirank

48 Ufree

49 Diriendly
50 Dirightened
51 ijurious
52 [jlively

53 [jgentle

54 Dglad

55 [jgloomy
56 Dgood

57 [jgood-natured
58 [jgrim

59 Cl happy

60 Dhealthy
61 Dhapeless
62 D hostile

63 Climpatient
64 [j incensed
65 D indignant
66 D inspired
67 [jinterested
68 Dirritated
69 D jealous
70 D joyful

71 Didndly

72 Blonely

73 Dlost

74 Dloving

75 Dlow

76 Dlucky

77 [jihad

78 C] mean

79 [j meek

80 ijerry

81 Dmild

82 Dmiserable
83 [:jnervous
84 [j obliging
85 Doifended
86 Qoutraged
87 [jpanicky
88 :patient

89 [:1 peaceful
90 C] pleased
91 [j pleasant
92 C] polite

93 D powerful
94 C] quiet

95 D reckless
96 D rejected
97 D rough

93 C] sad

99 [j safe

100 D satisfied
101 D secure
102 C] shaky

103 D shy

104 D soothed
105 D steady
106 [:1 stubborn
107 C] stormy
108 [:1 strong
109 C] suffering
110 C] sullen
111 D sunk

112 [:1 sympathetic
113 U tame

114 C] tender
115 [j tense

116 CI terrible
117 E] terrified
118 C] thoughtful
119 D timid

120 E] tormented
121 D understanding
122 D unhappy
123 C] unsociable
124 C] upset

125 C] vexed
126 D warm
127 D whole
128 D wild

129 [j willful
130 C] wilted
131 [:1 worrying
132 (3 young

APPENDIX E

 

 

 

k D r I lnv nt
- I -
Date
Name: Marital Status: __ Age: __ Sex:
Occupation: Education:

 

 

This questionnaire consists of 21 groups of statements. After reading each group of statements carefully.
circle the number (0. 1. 2 or 3) next to the one statement in each group which best describes the way you
have been feeling the past week. including today. If several statements within a group seem to apply equally
well circleeachone. Besuretoreadallthestatementsineachgroupbeforemaldngyourchoice.

 

 

 

 

 

 

129

I 0 Idonot feel sad. ' ° Idon't feelIamanyworsethan
' I feel sad. any body PM '
3 I am sad all the time and I can't snap out of it. ' (13mm? of myself for my weaknesses
3 Iamsosadorunhappy thatlcan‘t stand it. 2 Iblamemyselfallthe timeformyfaults.
3 I blame myself for everything had
2 0 :mt particularly discouraged about the that happens.
i I feel discouraged about the future. . - -
2 I feel I have no , to look fo to 9 0 I don t have any thoughts of killing myself.
8 ! Ihavethou htscfkiliing' ' m lf.butI
a I feel that thetfuture is hopeless and that would not carry them out. yse
“mm “pm“ 2 I would like toldllmyself.
3 0 Idonotfeellikeafailure. 3 Iwouldkillmyselfiflhadthechance.
I I feel I have failed more than the
average person. 1' 0 I don't cry any more than usual.
2 Asllookbackonmylife.allIcanseeis ’ Icrymorenowthanlusedto.
a lot of failures. 2 I cry all the time now.
3 I 1901 1311130013910” failureasaperson. 3 Iused to beagle tocry. butnowlcan't cry
, even though want to.
4 0 Igetasmuch satisfactionoutofthingsasl
usedto. I ll 0 Iamno more irritated now thanIeveram.
‘ I don't enjoy mu ‘3' the way M to. l I get annoyed or irritated more easily than
2 I don't get real satisfaction out of anything I used to.
anymore. 2 I feel irritated all the time now.
3 I am dissatisfied or bored with everything. a I don't get irritated at all by the things that
used to irritate me.
5 0 I don't feel particularly guilty.
‘ 1 £001 guilty 3 800d Pu" 0‘ the time. ‘2 0 I have not lost interest in other people.
2 I {001 quite guilty 310“ 0‘ the time. i I am less interested in other people than
3 Ifeelguiltyallofthe time. Iusedtobe.
2 Ihave lostmost of my interestin
l o Idon't feelI am being punished. “h" P°°P’°- . .
I I feel I may be punished. a I have lost all of my interest in other maple.
2 I expect to be punished.
. . ' ‘ u
3 I feel I am being punished. i3 0 §mafim°m about as we as
7 o I don't feel disap' pointed in myself. ' lat-igfémalung decisions more than
. I amdiuppointedin my-elf- 2 Ihave greater difficultyin making
2 I am disgusted with myself. decisions than before.
3 I hate myself. 3 I can't make decisions at all anymore.
Subtotal Page 1 CONTINUED ON BACK
"XTHE PSiCHOLOGicAL CORPOR ATIOX
‘i’ HARCOL RT BRACE 1m Axox 104 list
Copyright: 1978 by Aaron T Beck. All rights reserved Printed in the U S A
NOTICE: It is against the law to photocopy or otherwise reproduce
this questionnaire without the publisher‘s written permisson. 9.013359

 

 

130

 

 

 

 

 

 

 

 

 

 

M 0 I don't feel I look any worse than I used to. 1' 0 I haven't lost much weight. if any. Lately.
' I am worried that I am looking old or I I have lost more than 5 pounds.
a Imfmtluthawtlt‘: ban 2 I have lost more than 1 0 pounds.

_ ee ere are permanent c ges .

m my ap . ce that l e me look J I have lost more than 15 pounds.

unattractive.

3 I believe that I look ugly. I am purposely trying to lose weight by
eating less. Yes No
15 0 I can work about as well as before. In .
' It takes an extra effort to get started at 0 {81:22:31.8 won-led “b0“: my health
dang something I ed bo 'cal bl
2 mavemuanmyaeuvemmao ‘ mungmmifmpgs.“
a “M' stomach; or constipation.

I can t do any work at all. 2 I am very worried about physical
problems and it‘s hard to think of
much else.

3 o flee ‘4 I am so worried about m physical
1 I can. p as well as usual. problems that I cannot think about
I IdontsleepasweilasIusedto. “M3139.
2 I wake up 1-2 hours earlier than usual
and find it hard to get back to sleep.
3 I wake up several hours earlier than I .

used to and cannot get back to sleep. 21 ° .1 have 90‘ “on?“ any mu“ m

in my interest in sex.
1 Iamlessinterestedinsexthanl used
to be.
‘7 0 I don't get more tired than usual. 2 I am much less interested in sex now.
' I get tired more easily than I “lied to. a I have lost interest in sex completely.
J I get tired from doing almost anything.
=1 I am too tired to do anything.
1' 0 My appetite is no worse than usual.
1 Myappetiteisnotasgoodasitusedtobe.
'4 My appetite is much worse now.
‘J I have no appetite at all anymore.
Subtotal Page 2
Subtotal Page 1
_ '1th Score

 

TPC 0520-001 13 14 15 16 17 10 10 N

BCDE

 

 

APPENDIX F

BkHI l

 

STUDENT ID DATE

 

Beck, Weissman, Lester, and Trexler
HS Scale

Your answers to the items on this inventory are to be recorded on
a separate answer sheet which is attached. Print your name and
student ID! on the answer sheet, then finish reading these
directions.

Read each statement and then blacken the appropriate number on
the answer sheet to indicate whether the statement is 3:33 or
figlgg when applied to you. Use the following scale:

1. True

2. False

This is a questionnaire to find out the way in which you view
your expectancies for the future. Each item is a statement about
the future. Please read each item and determine whether it is
TRUE or FALSE as applied to you. This is a measure of personal
belief: obviously there are no right or wrong answers.

 

22. I look forward to the future with hope and enthusiasm.

23. I might as well give up because I can't make things better
for myself.

24. When things are going badly, I am helped by knowing they
can't stay that way forever.

25. I can't imagine what my life would be like in 10 years.

26. I have enough time to accomplish the things I most want to
do.

27. In the future, I expect to succeed in what concerns me most.
28. My future seems dark to me.

29. I expect to get more of the good things in life than the
average person.

(Continued on next page)

131

30.

31.
32.

33.
34.

35.
36.
37.
38.

39.
40.
41.

132

I just don't get the breaks, and there's no reason to
believe I will in the future.

My past experiences have prepared me well for my future.

All I can see ahead of me is unpleasantness rather than
pleasantness.

I don't expect to get what I really want.

When I look ahead to the future, I expect I will be happier
than I am now.

Things just won't work out the way I want them to.
I have great faith in the future.
I never get what I want so it's foolish to want anything.

It is very unlikely that I will get any real satisfaction in
the future.

The future seems vague and uncertain to me.
I can look forward to more good times than bad times.

There's no use in really trying to get something I want
because I probably won't get it.

 

APPENDIX G

Llf vnln

 

STUDENT ID’ DATE

 

The Life Experiences Inventory

Listed on the following pages are a number of events which sometimes
bring about changes in the lives of those who experience them and
which necessitate social readjustment.

ease se ve ts whic av ie t
gix_m9g§h_. To do this, blacken in the appropriate number on the
answer sheet from I to 7 to indicate the extent to which you viewed
the event as having either a positive or negative impact on your life
at the time the event occurred. That is, indicate the type and extent
of impact that the event had. A rating of 1 would indicate an
extremely negative impact. A rating of 1 would indicate an extremely
positive impact. (See example below)

 

 

e se d' ate those v ts w v
t s which but
gate thgge events. To do this, blacken in the number 19 on the answer
sheet. (See example below).
1 2 3 4 5 6 7 8 9 10
l. - extremely negative ----- > e O O O o o O O o O
. l 2 3 4 5 6 7 8 9 10
2. - moderately negative ----> O O O O o O O O O O .
12345678910 '
3. = somewhat negative ------ > O O O O O O O O O O
l 2 3 4 5 6 7 8 9‘10
4. - no impact > O O O O O O O O O O
1 2 3 4 5 6 7 8 9 10
5. a slightly positive ------ > O O O O O O O O O O
l 2 3 4 5 6 7 8 9 10
6. - moderately positive ---—> O o O O O O O O O O
l 2 3 4 5 6 7 8 9 lo
7. a extremely positive ----- > o o O O O O O O O O
l 2 3 4 5 6 7 8 9 lO
10.3 v OT e -> O O O O O O O O O O
vent ' th ‘

mm
W

49. Marriage

50. Detention in jail or comparable institution.

51. Death of a spouse

52. Major change in sleeping habits (much more or less sleep)

Death of a close family member:
53. Mother
54. rather
55. Brother
56. Sister
57. Grandmother
58. Grandfather
59. Other (specify)

 

133

60.
61.
62.
63.
64.
65.

66.

67.

2. I
3. I
4. I
5. I

7. I

10.I

134

 

l 2 3 4 5 6 7 8 9 10

extremely negative ----- > O O O O O O O O O O
1 2 3 4 5 6 7 8 9 lO

moderately negative ----> O O O O O O O O O O
l 2 3 4 S 6 7 8 9 10

somewhat negative ------ > O O O O O O O O O O
l 2 3 4 5 6 7 8 9 10

no impact > O O O O O O O O O O
l 2 3 4 5 6 7 8 9 10

slightly positive ------ > O O O O O O O O O O
l 2 3 4 S 6 7 8 9 10

moderately positive ----> O O O O O O O O O O
l 2 3 4 5 6 7 8 9 10

extremely positive ----- > O O O O O O O O O O
1 2 3 4 5 6 7 8 9 10

ve NOT e ri c d -> O O O O O O O O O O

this even; in the 13:;

six months 9; L; is ngt
W

Major change in eating habits (much more or much less food intake)
Foreclosure on mortgage or loan

Death of a close friend

Outstanding personal achievement

Minor law violations (traffic tickets, disturbing the peace, etc.)
M§l_: Wife/girlfriend's pregnancy

Egmale: Pregnancy

Changed work situation (different work responsibility, major
change in working conditions, working hours, etc.)

New

Serious

76.
77.

78.
79.

80.
81.

68.
69.
700
71.
72.
73.
74.
75.

Job

illness or injury of close family member:
Father

Mother

Sister

Brother
Grandfather
Grandmother
Spouse

Other (specify)

 

Sexual difficulties

Trouble with employer (in danger of losing job, being suspended,
demoted, etc.)

Trouble with in-laws

Major change in financial status (a lot better off or a lot
worse off)

Major change in closeness of family members (increased or
decreased closeness)

Gaining a new family member (through birth, adoption, family
member moving in, etc.)

_ m!

 

135

 

l 2 3 4 5 6 7 8 9 10
1. I extremely negative ----- > O O O O O O O O O O
l 2 3 4 S 6 7 8 9 10
2. I moderately negative ----> O O O O O O O O O O
l 2 3 4 5 6 7 8 9 10
3. I somewhat negative ------ > O O O O O O O O O O
1 2 3 4 5 6 7 8 9 10
4. I no impact > O O O O O O O O O O
1 2 3 4 S 6 7 8 9 10
5. I slightly positive ------ > O O O O O O O O O O
l 2 3 4 5 6 7 8 9 10
6. I moderately positive --—-> O O O O O O O O O O
l 2 3 4 5 6 7 8 9 10
7. I extremely positive ----- > O O O O O O O O O O
1 2 3 4 S 6 7 8 9 10
10.I ve NO e r en ed -> O O O O O O O O O O
thig event in the lag;
mont s o ’ ot
AW

82. Change of residence

83. Marital separation from mate (due to conflict)

84. Major change in church activities (increased or decreased
attendance)

85. Marital reconciliation with mate

86. Major change in number of arguments with spouse (a lot more
or a lot less arguments)

87. naggigg_mg1g: Change in wife's work outside the home
(beginning work, ceasing work, changing to a new job, etc.)
Married female: Change in husband's work (loss of job,
beginning new job, retirement, etc.)

88. Major change in usual type and/or amount of recreation

89. Borrowing more than $10,000 (buying home, business, etc.)

90. Borrowing less than $10,000 (buying car, TV, getting school
loan, etc.)

91. Being fired from job

92. £31.: Wife/girlfriend having abortion
figmglg: Having abortion

93. Major personal illness or injury

94. Major change in social activities, e.g. parties, movies,
visiting (increased or decreased participation)

95. Major change in living conditions of family (building new
home, remodeling, deterioration of home, neighborhood, etc.)

96. Divorce

97. Serious injury or illness of a close friend

98. Retirement from work

99. Son or daughter leaves home (due to marriage, college, etc.)

100. Ending of formal schooling

 

136

 

l 2 3 4 S 6 7 8 9
l. I extremely negative ----- > O O O O O O O O O
l 2 3 4 S 6 7 8 9
2. I moderately negative ----> O O O O O O O O O
1 2 3 4 5 6 7 8 9
3. I somewhat negative ------ > O O O O O O O O O
1 2 3 4 5 6 7 8 9
4. I no impact > O O O O O O O O O
1 2 3 4 5 6 7 8 9
S. I slightly positive ------ > O O O O O O O O O
1 2 3 4 5 6 7 8 9
6. I moderately positive ----> O O O O O O O O O
l 2 3 4 5 6 7 8 9
7. I extremely positive ----- > O O O O O O O O O
1 2 3 4 5 6 7 8 9
10.I ave NOT e e d -> O O O O O O O O O
s eve t n t at
s mo t s r t
Wee
101. Separation from spouse (due to work, travel, etc.)
102. Engagement
103. Breaking up with boyfriend/girlfriend
104. Leaving home for the first time
105. Reconciliation with boyfriend/girlfriend

Other recent experiences which have had an impact

List and rate.

-109. Beginning a new school experience at a higher academic level
(college, graduate school, professional school, etc.

110.

111.
112.
113.
114.
115.
116.
117.
118.

106.

10
10
10
10
10
10
10
10

on your life.

 

107.

 

108.

 

Changing to a new school at same academic level
(undergraduate, graduate, etc.)

Academic probation

Being dismissed from dormitory or other residence
Failing an important exam

Changing a major

Failing a course

Dropping a course

Joining a fraternity/sorority

Financial problems concerning school (in danger of not having

sufficient money to continue

)

'k

‘If'; 1.!

 

 

APPENDIX H

II_ tin hIr'Th M '

SOLICITATION SPEECH TO BE READ TO STUDENTS IN PARTICIPATING
CLASSES PRIOR TO THEIR COMPLETING CONSENT FORMS

THE HOOD STUDY

Your instructor has consented to let us conduct a study in this
class. First, we would like to tell you a little about the study, and.what
your participation in the study will involve. Then we will pass out
consent forms for each of you to complete should you agree to participate
in the study. Individual participation in this study is voluntary.

This study will require that you complete a questionnaire in class
several times along the course of the term. The questionnaire will be
given often, but not in every class. You will 39; be told in advance when
the questionnaire will or will not be given. The questionnaire is a
measure of affective state, and we will be looking at group affective
states over time. A handout explaining the study in more detail will be
provided shortly after you complete the last questionnaire in this study.
Data from this study may be combined with data from other studies in this
or other classes.

Participants will be expected to complete the questionnaire eyegy
time it is given in class. However, should you miss a class when the

questionnaire was given, on s 0 st co t nu o c le t
e t ai e eac e t i iven w n on are in as. a tic a s
- dwas o ten-s 0114 ewe 's-ivenwhe 1: :e_,
1 ss. Never ski it st aus ou hav isse li t on her
t s b 'v .

---> Are there any questions at this time?"

I
I
1 We will now pass out consent forms for this study. You

1 should read them thoroughly. Then, if you agree to participate
1 in the study, please sign the form, and put your Student ID#

1 and today's date on the lines provided.

1

I

I

I

Thank you very much for your time and cooperation.

--> ** If any questions are asked by the students, the following
procedure should be followed.

1. If a student's question can be answered by repeating one or more
of the sentences from the speech above, then those sentences will be
reread in response to the question.

2. If a student's question cannot be answered by repeating one or
more of the sentences from the speech above, then the following reply'will
be given to the student's question:

We cannot answer that question now because it may interfere with the
study. However, once the study is completed, we will be happy to answer
that or any question about the study.

137

 

 

APPENDIX l

IIIttln h fth "P I'll I"

SOLICITATION SPEECN TO BE READ TO STUDENTS IN PARTICIPATING
CLASSES PRIOR TO THEIR COMPLETING CONSENT FORMS

PERSONALITY STYLE

Your teacher has consented to let us conduct a study in this class.
First, we would like to tell you a little about the study, and what your
participation in the study will involve. Then we will pass out consent
forms for each of you to complete should you agree to participate in the
study. Individual participation in this study is voluntary.

This study will require that you complete four questionnaires in an
out of class setting, and three questionnaires in class. To complete the
out of class questionnaires, you will need to sign up for a 30 minute
questionnaire session on one of the sign up sheets posted in the back of
the classroom. A wide variety of times are available for those
questionnaire sessions to make that part of the study as convenient for
you as possible. The first of the measures to be completed in class will
be given today. The second and third in-class questionnaires will be
given at a different time later in the term. The questionnaires measure
various aspects of personality and style. The experimenter will compare
group results for relationships between these measures. A handout
explaining the study in more detail will be provided shortly after you
complete the last questionnaire in this study. Data from this study may be
combined with data from other studies in this or other classes.

---> Are there any questions at this time?“

We will now pass out consent forms for this study. You

1 should read them thoroughly. Then, if you agree to participate
in the study, please sign the form, and put your Student ID! and
today's date on the lines provided.

 

Thank you very much for your time and cooperation.

---> ** If any questions are asked by the students, the following
procedure should be followed:

1. If a student's question can be answered by repeating one or
more of the sentences from the speech above, then those sentences will be
reread in response to the question.

2. If a student's question cannot be answered by repeating one
or more of the sentences from the speech above, then the following reply
will be given to the student's question:

We cannot answer that question now because it may interfere with the

study. However, once the study is completed, we will be happy to answer
that or any question about the study.

138

 

 

APPENDIX J

D n In lnfrmlan

DEBRIEEING INFORMATION FOR PARTICIPANTS IN STUDIES CONDUCTED
8? MARK H. WAGNER IN SPRING OF 1988

This handout is for students who participated in QLEDEI or teth of
the studies (The Mood Study, and Personality and Style) conducted.by Mark
H. Wagner in Dr. Lindel's PSY 215 class or Leslie Wolowitz's PSY 160 class
in Spring of 1988. As each of these studies were part of a larger study,
the larger study will be explained here.

W
The purpose of this study was to to examine the validity of
F RESSI (HTD). This theory suggests that some
people have a way of thinking that makes them susceptible to becoming
depressed (i.e. hopelessness depression, one of many types of depression).

This way of thinking is called a depressogenic ettgibutiogal etyle (DAS)
because these people have a tendency to attgibute the cause of important

negative life events to gauges that are internel, globe}, and etetle. For
example, if someone attributed being rejected by a member of the opposite
sex to their not being attractive to any members of the opposite sex, they
would be making a depressogenic attribution. It attributes the cause of
the event (being rejected by a.member of the opposite sex) to a cause that
is internal (being unattractive is a personal characteristic), global
(feeling unattractive to egg member of the opposite sex, not just the one
who rejected them), and stable (usually one's attractiveness does not
change very quickly).

According to HTD, it takes more than a DAS for someone to become
depressed. One first needs to experience an_imeettent_neeetige_1i1e_exent
(INLE) to make depressogenic attributions about. That is, an INLE must
happen to a person with a DAS to bring about a hopelessness depression in
that person.

HTD also suggests that the relationship between a DAS and the onset
of hopelessness depression in the presence of an INLE is meeieteg by the
actual attributions one makes about the cause of that event. Recall that
a DAS is only a tengegey to make depressogenic attributions. HTD suggests
that one will develop hopelessness depression

We].
WW (internal, global and stable) about the cause of
an INLE.

So the chain of events that typically lead to the development of-

hopelessness depression are as follows: First, a person has a DAS. Second,
that person experiences an INLE. Third, the person actually attributes the
cause of that INLE to internal, global and stable causes. Fourth, making
those attributions makes that person feel hopeless and thereby brings
about hopelessness depression.

W
First, it is important to understand that both of the studies
conducted by Mark H. Wagner in this class were part of the same test of
HTD. This study was presented to you as two separate studies to make it
less likely that you would guess what the study was testing, which might
have biased the data.

139

 

 

140

The questionnaires that were used during the 30 minute questionnaire
session that took place out of your class measured depressive symptoms,
feelings of hopelessness, DAS, and recent INLEs you may have experienced.
The questionnaire that you completed repeated times in class was a measure
of transient depressive mood (TDM). The questionnaire that you completed
in class at the beginning of the term measured yee;_egn_inte;gel_eteegetg
for what you would consider a good or a bad grade on your first midterm
exam. You were not told that the last measure of TDM was going to be two
days after receiving your midterm exam grades so you would not suspect
that the study was measuring your moods around receiving your grades.

The following is an explanation of this study's test of the DAS-INLE
interaction in causing hopelessness depression. Students data will be
divided into the following groups:

A low exam grade (relative to your own standard) is assumed to be an
INLE. It is predicted that only group number 1 will show a lasting (2
days) increase in depressivermood in response to the receipt of their exam
gra es.

The following is an explanation of the study's test of the the
mediation role actual attributions play in how DAS and INLBs interact to
cause hopelessness depression. After receiving your exam grades, you
completed a questionnaire that measured the actual attributions you made
regarding why you performed the way you did on the exam. It is predicted
that those who had DAS at baseline made depressogenic attributions for how
they did on their exam. It is also predicted that these actual
attributions will correlate positively with increases in depressive mood
for those who received a low exam grade. Finally, it is predicted that
students actual attributions regarding their exam performance will better
predict their changes in TDM than will their baseline measures of DAS.

The last questionnaire given to you was a measure of hopelessness.
HTD suggests that hopelessness depression is ultimately due to the
development of feelings of hopelessness. Therefore, it is predicted that
only students from group number 1 (above) will show elevated feelings of
hopelessness from baseline to this last measure. ‘

Finally, it is important to note that DAS, depressive mood, and
feelings of hopelessness occur in varying degrees and might change over
time. Therefore, if your responses on measures used in this study
suggested mood change, it does get necessarily mean that your mood change
warranted any concern since most peoples' moods vary from day to day. If
however, you are concerned about any feelings you think may have been
precipitated by this study, please contact Mark a. Wagner through the
graduate office of Psychology at Snyder Hall (355-9561) or Dozier W.
Thornton through the Psychology office at R.B. Olds Hall (353-3249).

I want to thank you for participating in my study. If you have any
further questions about this study, or are interested in learning the
results of the study, feel free to contact me through that graduate office
of Psychology in Snyder Ball.

 

 

I!

 

 

.codusaouuou Asauusm ooussoe I_~um .co«usaeuuou «squash I am .ouusausue u cs>ao ecu osds> a I u .OAm

Iowa-«usue h co>ue new made» a I m .odu .uee oz» ecu mm :4 uceeeuoca I_~ue In: o>49susEso Item .850 .>m¢ oEse
an» unseen aura 00:40 .N oubsh non as mass on» ad on sense new use» use .codusEuOuca sue» eoueoun Doc oao

Ahmad.

ecu soasusbse

.us be hueasue: oessoo
uaasboHO one

M.

«senses see :0
emanates 0:» co seeouuso yceeo>oueus o>qusooc new seasoebss

souooe uo omsuo>s I 04m ocs

ozone uoc nu ma ounce new use» one .ouascc04ueeso occuuaeauuuc undocuausm or»
.ouuscco«usoso euaum «occausbuuuuc
headseouo ocs auudabsum 0:» co souooo uo emsue>s

 

 

 

 

 

I one Icesu>smlueae xuozu s>au00no< uoouu< sueuuusz or» so once. I m:ao¢<x .eocsause usuuuen no see>mscs
OCdfip—OQOOHHOO 0:8 09:.“ VOHODGO av». glad O8» I63 930—; naval :Odulfluuou 0:» IO» Ufluufluou Dun—6h. SOUO HOb .
«e. e«.1 a~.n as. on. en. en. assessoeoeos :4 Issues I eusoecr meanness In
no. e~.s «H.n an. oo.~ so. an. assessouoeo: sq cusses I «neocer oseeossn on
so. Ie.u «H.n so. so. so. es. Iaosssseoeoz n uses I eueossz oseeosss me
66. p6.: ne.n so. me. so. no. sass-Ioeseoz n cede I zueoear oseeosss I"
on. mm.: «a no. o~.«u one I zaaoeer seeds-en
on. no. as ea. mm.u . can I mneoe<a scansesm
on. no. «AIm so. on.n an. no. you me
no. an. an an. no.~ and I «saved: scansesm
ea. an.: an en. e~.~n are I nuance: Isaac-Is
an. «n. ne.m on. on. on. In. use «a
so. no.1 n~.n no. me. 66. cu. gas I xueoeer seduces: an
as. 46.: n«.n ea. oo. so. no. ace I m-ao¢<r sequence on
no. he. u«.n me. no. no. on. out I enaoeer oseeossn o
«6. ea. ««.n 6.. as. no. on. one I sneo¢<r seeds-an n
no. on. n«.n on. an. no. es. on: I «neuter seduces: I
do. ho. n«.n on. so. no. no. one I «never: seduces: n
do. «n.: an «a. no.1 osouuso seen I one I sunset: seeds-en
«6. un.: an we. ep.: one I «never: sedge-en
no. mo. an mm. mm. snouuso sewn I unauttz sequessm
no. on. mn.e es. en. «0. an. no» a
do. o~.u an on. em.u seouuso sane I one I «never: Isaac-an
«o. I”. an «I. do. one I snooze: oseeossn
oo. «c. mm no. mo. seoousO lsxa I zuaoemz seducesm
«a. an. mn.e «a. nn.a on. on. use a
use an up .In. encuoqosue .Ie. use ecu use In codussvs sea ecu subsqusb ucsocsesocd sense
a uosacazuaz h «a ca .890 goes sud: susuus>oo or» no sausCOum .uo
.o«m ecu a .04m possessed es» «0 >uucs ssoscsunsE«m .use :4 seabedoeum
use I
I I- (I. . O. .. .u ...OO. .4 .O -OflOuiutae . I . s-
e cease

 

K Nufllflhh‘

 

.141

 

.—0. v e I II .mo. v e I .esusuw ucs>u spy; s>Iusmsz c. :ovusvte> o: e. store .coquO—se ea sso .egucoE eve «use sea r. ucvtsuuo eucs>s see, s>puhhkk
h—sEstuxs or usutoest newsflasm .Lsxop to scvr a. stove htoucs>cu cove-steso seen .speise cut-sees vsuus—se h—psv era to tease a. cast eve» .nw I e .euoz

 

so." Isqum IIIIL
Ie.s I>IIIeIz

htoucs>cu
co." «0.- Ion. Ion. on.. so.- an. Inn. on. «N. nu. covessuuum
success—see: e.

on." .00.. Icon.. «0.: v0.0 n—. «a. In. ~n.n an. Oucssu n OE,»
cu scI—sesn

 

 

 

co.” seem. m«.: an. Nu. ON. ON. 00. , an. eases-Opens:
n OI.»

 

oo.— 00.- On. 00. Ma. an. Inn. «0. esseeeO—anz
s:.—sosn

 

sosta
so." .Iae. -.- as. no.. «a. ne.- IIIu
IIIIuInso

sseouao
so." .Iuo.- «a. n_.- IIee. ee. IIIu
IIIIoIqssm
sueleoILsa

00." «an. on. scan. 00.0 [saw so»
sovuanotvud

aesls>s. o(
00.” no. No. no. LO» On an
pscovusavLau(

sepeesteso
oo.~ Io.u IIoe. «140((1

 

 

 

 

 

Oo.~ 00. c:au¢¢t

 

revuestaso
co." u-40((z
seI—ssse

 

seesaw area essceespseo: ease-esp users-op spate slouuso surslteu ussIs>s gut .eso .aso .eso

ucs>u nes>c~ c. secseu uses: uses: [can seem Itsa [new to» s. an cuau((t unaudix cuau((x

sv. .aso n si.» e» n slwp scI— s> u s>.u to» so.» pure.» n st.» N sl.h se.psesn
s>vus sz dose ss.pseso use-o -os no -usnesm usevtuue usavtuu<

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

.— 5:22.:

142

 
 

.no. v a I I .mo. v a I I .sEbouso esuw e>.uusnbsm c. to .esueum acs>u seve s>Iusmsz r. covusvts>
o: I. store .sovuuspse ou sso .svetu Leora use: Isaac; use: maven seasons; euusneam .ezucoe III «use see c. ocvtsuuo euce>s svvp s>Iusmsc a—sesnwnn
o: usutoast uuoenbsm .tszo— to It's I. stone toacs>c~ soesostaso «use .s—eEse rot-sect nevus—so Apps. sea assess to sense I. ease eve» .n— I c .suoz

 

co." . Isa-um ucs>u
III; IIIIIsIz

 

htoucs>rn
so." -.- Is. oe. n~.- m~.- ”N. «e.. "I. as. IIIIIIumum

success—sac: c.

oo.~ Inn. IIOD.I m~.- 0n. 0". do. nn.: Nu. Our-cu n OEub
ea stereo-o

 

 

co." ow. mu.u an. on. on.a no. oo. essence—see:
n ssvb

 

oo.n «0.- nn.u on. o~.I av. an.n eeOreeO—Onoz
ssI—sesm

 

state
so." no. ee.- an. o—.- e~.- IIIu
_ I>.IoInso

sisouso

co." Esxu
IIIIoInss»

socsltoetse
o~.I Ia¢.I vn.I stu Lev
cocuaa—Luu(

 

 

143
8.
is

 

aesis>e. 91
co." co. on.I IO.I LO» on an
. pseovuanvLu01

so-sssteso
co." «o. «u. c-40((x

n st.»
rev-esteso
oo.~ eo.- cuau<gt

N at.»
rev-esteso

DO." ¢I40((!
ssI—sosn

 

 

 

 

Isa-um Atou eessees—sao: eescees— ease-es. sestu secuuso corsets» uss£s>sceo( .eso .eso .aso

IIIIu -IIIIS I. III-Io -Isoz -Iso: IIIu IIIu -IIe seem III Ipaum c-eu¢¢: a-eu¢¢r ¢-eo¢<r

I». .eIo n It.» II n II.» II.. II I II.» to. IIII .IIIII a II.» a IIII I:..III.
IIIII I: III. II..III¢ -III. -uI so -oInesm -ss.taa< -se.tII<

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

u 023

144

to.ue—Ls> or a. stseu .covuos—se ea sso .sustu t.sru ea.) unease:
o: esutoaet sausnesm .tsze. to s:.s I. stoua utenss>ru ropeestaeo sues .s—aEII gut-sees oeuospse >——:» or» usage. to use-n I. seen e.gp .ON I : .suoz

. o. v a I II .mo. v e I I .ssoouso Esxu s>vuusnnsm c. to .esusum uee>u s~.w s>.usmsz c.
s

c.sb usutoast sausnesm .egucei x.e case or» :. uc.tsuoo eucs>s sII— s>vus

c A—sesths

 

oo.~

sea-um aes>u
I..e II..IeIz

 

an.

Inc.

00.

hN.

In.

Ice.

«N.

ON.

misuse):—
ce.eesteso
Jose

 

sh‘.

no.u

nu.

mm.

mm.

Cn.I

ease-es—sao: :.
sec-go n sE.b
ea sc.—sesn

 

an.

0N.

INC.

an.

nascees—seo:
n st.»

 

sN.o

«N.

an.

ON.

Nu.

an.

gaseous—see:
0:.psese

 

Nu.

an.

sesto
lsxu
II.IuInso

 

sleousO
lean
II.IoIqsIm

 

sOQ.

Nu.-

sueslteytse
lsxu use
cocuaavhuu(

 

.0.

an.

utensII. u<
LO» O— an
peso—uanwsuv<

 

on.

revuessaso
u-4u<(x
n s1.»

 

«N.

so.eesteso

 

co."

so.esstaeo
uceu<et
sc..seee

 

Isa-um
aes>u
II.

O>uue O2

 

 

stow
Iss>cu

III.

 

esscees—sao:

e. surseu
n slob on
scI—seeo

 

OOO‘OOO—
I030:
n s!.»

 

OOOCIOO—
0.30:
I:..
I0...

sense
[new
0) a
use no

 

seeouao
Issu
II.»

-oInsIm

 

 

soc-Ito.
-Lsa [sum
to» so.»
nanotuu(

 

usele>s.so(
to» sphun
pece.a
uaautuu<

 

sga
CIAU<¢Z
n OIvh

Ina
¢I40((I
N Olvh

 

 

0‘8
CIJU<(!
s:.peeso

 

 

 

. .uuec esxm «I a.mee e vs: sea a a muce> o .4 s
as e

A.

et.ea son 3.

a «use a I

.gio“. . v

a open»

..u~ A.

.eu :
s> .

me my. nu

s can teues> u
e s s so

.no. v a I II .no. v a I I .eeeoetsn s>.u.tse500 to» use.>oue
a. sane e.g. .AspaEsI got-sees nevus... A..s. Ira o» eseoaao as. ease sus.eEOo to» nevus—ea spas-e avenues ..s. sea to ecu-e e. ease e.eh .oo. I I .ruoz

145

 

 

 

 

 

 

 

 

 

 

 

 

 

 

co.— IvN. 0.. e.. .n. on. Ma. No.- In.. IIoN. IIeN. Nn. Isa-um ues>u
I..4 II.quIz
Atou¢s>cu
8." DH... ssOD. ssoo. N~II '0. 0°. sen'. ssOMI ssNfl. ssfln. COIOOOIJ—Mnm
success—see: c.
00.. IeN. IInn.- an. no.1 no. nu. eo. 00.- «0.- set-Io n IE.»
on s:..ssen
oo.. IIeo. no. n.. no. IION. IIen. IIeN. IIsn. Incense—see:
M II.»
oo.. no.1 «a. no. ION. IInn. IIen. IIoe. nest-es.sao:
sc..sese
seeto
oo.~ IIue. InN.n no. a..- on. n..- lsxu
II..oInso
sleuuso
co.“ IInm.u no. I... IIeN. e.. stu
II.IoInIIm
. sues-totes
oo.~ IInn. eo. e..- N..- seen to.
ce.use.tua<
uesls>s. o<
oo.. o.. IINn. us. to. s. an
.eco.usa.tuu<
:o.eesteso
00.. 0.. IIne. cueueex
n sl.p
so.essteso
oo.. IIOn. ¢I4u(¢z
N sl..
so.eesteso
00.. c-au¢¢x
s:..sese
nausea Assn essence—see: eeseeus. «Issues. saute seoouso cocaine. assis>s.gu< .aso .eso .aso
«III» -IIII. I. III-Io -Ieo: -Ieo: IIIu IIIu .III III» II. I.»Im I-suIIr I-IuI¢r I-4uI¢x
s». .aso n st.» ea n st.» sI.. I) a II.» to. so.» .s:o.u n es.» N st.» se..sese
s>.us suso ss..seso use-o use no uoeneam use.suu< use.tuu< .

 

 

 

 

 

 

 

 

 

 

 

 

 

 

m open.

 

146

s>.u.tsseoo to. vso.>0ta I. ease I.I. .sEbuuso Esxu s>.uueneam I. Io.Is.ts> or I.
qusnasm .Ae.eEII rot-III; usuus.eI A..:. sea ou usIoaeo II. sues sus.s£oo to.

.no. v a I II .no. v a I I .IsIoetsn

etseu .Io.uus.sI on sea .sustu 5.0;» su.3 sane-e use:
tease—II s.aEII uuenaaI ..:. III to III-e I. Ives I.I. .nn I I .suoz

or.se usutoest

 

OO.~

ON.

0O.

NO.

nu.

"N.

ha.

an.

Isa-um acs>u
I..I II.IIIIz

 

0O."

an.n

IIno.

CC'FI

IN.I

so.

MN.

Inn.

htoucs>cu
Io.IIstasO
Jose

 

ssfln.o

«a.

On.

nu.

OO.I

IIIIIIs.saox I.
IIIIII I II.»
II st..sIIa

 

OO.«

CON”.

3..

nN.

IIon.

an.

IIerIs.seo:
n IE.»

 

OO.~

nn.o

Ma.

Ma.

ssOGI

IIseIIs.seo:
s:..sIso

 

sOGII

50.

on.c

Issue
luau
s>.uosneo

 

sleuuso
stu
II.IoInIII

 

OO.~

eoesELoILsa
seem to.
:O.u:n.suu(

 

I'M.

IIIEIII. 01
LOI s. an
pecovvan.huu(

 

OO.~

nu.

ee.IIstasO

 

s'fl.

 

0O.—

Io.IIstasO
euaocdt
II..IIIn

 

 

Isusum

aes>u

s».
s>.ue 0:

 

Ate»
oes>e~

IIII

 

IIsIIIspsee:
e. sueseu
n st.» so

s:..sIsn

 

IIscIIs.
uses:
n st.»

 

IIsIIIs.
uses:
It.—
III-n

 

sestO
[sum
s> a
-us so

 

sIOUIIO
lsuu
II.»

-oInsII

 

surelto.
Itsa Essa
to» so.u
0:3.Lau<

 

uIsss>s.gu<
toy sphum

.III.»
-II.III<

 

I‘S
CIJU((Z
n Ol.b

 

I‘a
¢I40((l
N OI.»

 

0‘8
CIJU((I
s:..sIse

 

 

 

m o.neh

 

 

. o. v a I II . o. v a I I .IsIoetaa
II e.gIsIou to. eIs.IoIe I. IIII I.I..IIou.Io IIII II.IoImsII I. II..I.III II I. IIIII .II.III.II I..Iso .IIII I.III II.I I eeII. II.II IIIIIIII

qus as» .AspeIsI sou-III; tacos—se A..:» sea ea osIoeao II ease sue—else to» tease—II epsIsI IosnaeI ..:. sea co esIsn I. ease I.I» .on I I .swoz

 

co.” ImN. ImN. IsN. In. no. a..- NN. Ien. IIee. no. Isa-um Ics>u
. I».4 s>.IIusz

 

Atouee>cn
OO.. 00.: IIoe. IInm. »n.- s.. IIno. nN. Ion. ON. Io.IIs“uum

 

IIscIIs.seo: I.

OO." Inn. ¢~.I «N. NO.I en. ON. an. OO. suessu n OI.»
II s:..sIsn

 

OO.” IIno. NO. nu. IIon. IIOO. IIOQ. an. eeeeeeO—Onox
n OI.»

 

OO.. OO.- mg. Inn. IIsn. IIno. nu. IIseIIs.seo:
er..sIsn

 

ensue
66.. ne.- «o. Is. no. I~.- IIII
II.IuIneo

 

sIeoIsO
ee.. IIII
II.IoIqIII

 

147

I soc-ltovtsa
OO.~ Icon. hO.I nO.I »O. Idem LO»
cO.u:n.Luu<

 

uIsIs>s. o<
OO.. s». Inn. ON. to» s. an
peco.v:n.suu<

 

ee.IIstesO
8." s8. s08. 8.3—0;
0 II.»

 

ce.IIsLesO
OO.. on. ¢IAO(¢:

 

. Io.IIsLasO
co." unbuedt
sI..sIse

 

 

Isa-um area IIsIIIs.sao: IIseIIs. IIsIIIs. state sloouso soc-Its. aesls>s.g0( .esO .esO .esO

IIIII -IIII. I. IIIIII -Ieoz -IIoz IIII IIII -III IIII to. I.I.I I-II¢¢x I-IuI¢r I-Iu<¢r

s». .aso M II.» ea n .I.» s:.. II I s>.u to. co.« .sro.» n ..I.» N sI.» sI..sIen
s>.us muse s:..sIso III-n nos 00 nosnaam Ise.tau< Ian.tuu<

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

a o—na»

 

m

 

.«o. v a I II .no. v a I I .IsIoetsa s>.u.tsesoo Lo» oeo.>eta I. sass I.e» .tsxo. to s:.: I. stouI stoute>t. to.IIetaub

soon .Ae.eEsI gusseIet nevus—II A..o» era 0» osIoaao II. suso sue—aEOU to» scans—II s.eIeI ausnesI ..a» sea Io osIIe I. ease I.I» .»o I I .euoz

 

 

 

 

 

 

 

 

 

 

 

 

 

 

oo.« II«n. o«. n«. no.- o«. N«. no.. .N. o«. IeN. ««. Isa-um Ite>u
I... II..IIIz
htoucs>cn
OO." «O. scan. stun. 8.0 0n.0 an. can”. such. ssNH. sNN. IO'OOOHM“
IIseIIe.sao: I.
oo.« IIoe. IIee.I no. «0.- no. »«. ««. N«.- «o. eac-Io M II.»
ea sc..sIso
oo.« IIom. no. No.. o«. IIeN. IINn. o«. o«. IIIIIIs.sao:
M II.»
oo.« co.- «o.- N«. v«. InN. IION. 0o. IIIIIIs»saoz
sI..sIso
sosto
oo.« IIoe. IoN.u »o. n«.- N«. »o.. stu
II..oI.so
sEouueo
oo.« IImm.u ««. I«N.u ImN. «o.u IIIN
II.IuInIII
soc-Ito.tsa
oo.« IIon. o«. »«.I no.. stu to.
I to.oan.LII<
aesss>s. o<
oo.« »o. IInn. N«. to. s» In
.Ico..:n.tuu<
to.IIstasO
oo.« *0. IIoN. «Ieu<¢z
n sI.»
co.IIsLesO
oo.« o«. enou<¢z
N sI.»
co.IIstesO
oo.« unauedt
st..sIsm
Isa-um Ito. IIscIIs.eaoz IIscIIs. IIsIIIs. Issue sesuuao eon-Ice. Icete>s.eo< .aso .asO .asO
IIIII -IIII. I. IIIIIu -Iee: -Ieox IIII III. .III IIII to. I.»II I-IuI¢r I-Iue¢r «.3614:
s». .aso n IS.» 0I M II.» or.» II I II.» to. co.» .eco.I n II.» N sI.» sc..sIeo
s>.II Ioso o:..eIIo -sIIo use so susnoam Iae.tII< use.iau<

 

 

 

 

 

 

 

 

 

 

 

 

 

= 02a»

 

 

 
 

.«0. v a I II .mo. v a I I .IIIongan I>.I.LInsou Lo. 0I0.>osn I. Iqu I.;» .LIzo.g so 0« I. IsooI h;0u:I>c« co.IIIL0ub
qun .AI.0EII goLIIIIL ququII A..:. Is. ou IIIonao II. III: III—0500 so. quoI.II IpaIII «nonnaI ..3. Is. :0 uIIIn I. IuIv I.:» .«N I 0 .I002

 

 

 

 

 

 

 

 

 

 

 

 

 

00.« o«. no. an. «N. m«. »0. »0. n0.n III. on. on. Isa-um ucI>u
I..4 II..IIIz
hsoucI>:«
00.« 0I.u IIII. IIsn. In.- «0.- N«. «I. IIoo. »«. IIno. :o.III“uum
IIIcIIopIao: c.
00.« ««. o«.- 0m. 00.- «0.- I«. »0. I0. o«.- Iuchu M II.»
II I:..IIIm
00.« IInu. on. In. 00.- I0. I0I. on. «I. IIIcIII«Ino:
n II.»
oo.« ««. on. n0.u «0.- III. »N. III. IIIcIIIpIno:
I:..III0
ItIsa
00.« «a. nu.. «0.- »~.I »n. o«.- IIIu
I>.IuInao
IIouuao
00.« IIn.u II»m.I I0. an. «I. IIIu
I>.uonaam
«u.
AW IUIIILIILII
4| 00.« II»I. o«.- «0.- In.- IIIu no.
co.Iaa.Luu<
IcIII>I. u<
00.« N0. o«. I«.- II» I. am
.Ico.u:n.suu<
co.IIILan
00.« «n. IIso. «IJUIII
a II.»
co.IIILnIo
oo.« o«. c-40(¢x
N II.»
co.IIILnI0
oo.« cuau<¢x
Ic..IIIn
Iauoan hsou IIIcIII«Iao: IIIcIII. IIIcIII. IuIsa IIouuao IucIILo. «IIII>I.¢o< .nIo .nIo .nIo
IIIII -II>I~ I. III-gu -Iaoz -Inoz IIIu IIIu -III IIIu so. I_III ¢-Ju¢¢: c-4u<¢x c-au¢¢z
I». .aI0 n II.» II 0 II.» I:.. I> I I>.I so. so.» .Ico.a I II.» N II.» I:..III0
I>.uI I: ono I:.«III0 -IIIo -oI no -oInnam -aa.sua< Ian.Luu<

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

« a «an...

150

 

.«0. v a I II .mc. v a I I .IIIIILII I>.u.LInEou Lo. uIv.>oLn I. Iqu I.:» .Isucos I.I III: In" :. 0:.Lauuo IacI>I I... I>.quIc ApIIILuII o: quLoaIL

IIUInaam .aopnsII suLIIIIL uIaoopII h«.a» Is. I» uIIonno II. Iqu III—anu Lo» quuopII I.

II qunIII ..I. III II III-I I. IIII I.I» .on I I .Iuoz

 

oo.«

Iaqum ucI>u
I... I>.uIIIz

 

o«.u

IInQ..

Conn.

o«.-

no.

IIO'.

so.

Io.

ALI»:I>:«
co.IIILnIo
goon

 

IINI.

IIOM.I

N0.I

««.

N«.

IOMII

IIIcIII»Inox :.
III-Io I Is.»
Ia I:.«III0

 

oo.«

Iva.

N«.

N«.

IIIIIII»Ino:
I II.»

 

Io.-

no.

no.

IIIcIIIpIao:
I:..III0

 

IINI.

IN.I

N«.I

IO.I

ItILo
IIIu
I>.Iuonao

 

co.

««.-

.C”.I

no.

IIIouao
IIIu
I>.qunIaI

 

oo.«

Inn.

II0¢II

IucIILIILIc
IIIu so.
co.u:n.Luo<

 

u:III>I. u<
Lo» I» an
plcowuanusuu(

 

Ion.

co.IIILnIo
acau¢¢t
M II.»

 

no.

:o.IIILnIo
c-40¢¢t
N II.»

 

oo.«

co.IIILan

CIJU(¢!
I:..III0

 

 

IIIIII
ucI>u
I..

I»... I:

 

use»
III>¢«

soon

 

IIIcIII.Inoz

:. Iochu
n II.» 0.
I:..III0

 

IIIIIII.
IInoz
M II.»

 

IIIcIII.
IInoz
I:..
IIIII

 

 

IIouuao
IIIu
I>.I

-IIIIII

 

IucIILo»
ILII IIIu
so. co.»
-II.IIII

 

ucIII>I.;o<
Io. I.aum
—OC0.0
-aI.IaI<

 

0‘8
CIAU¢¢I
n It.»

 

Ina
¢I40((!
N II.»

 

I‘a
BIJU¢¢I
I¢.«III0

 

 

 

 

0 Ipam»

151

 

.n . v a I I .IIIoasaa I>.u.LInEou Lo. vIv.>oLn I. Iqu I.&» .IsucoI x.I
IuoI nan .AI.0III gusIIIIL quuI.II A..:. III on uIIoano II. Iqu IuI.ano

 

.nO. V E I II

uIIa Is» :. 0:.Laouo ucI>I I... I>.quI: A.IIILIII Ico IIII. II quLonIL
Lo» quoI.II I.0III uoInnaI ..a. Is. In IIIIA I. III: I.¢» .on I I .Iuoz

 

 

oo.«

Iaqum aco>u
I... I>.IIIIz

 

oo.«

mu.-

I'm“-

«N.I

«o.

INn.

CC”.-

ICNnI

IIno.

ALI.:I>:.
co.IIILnIo
ono

 

oo.«

o«.

IIDHII

ON.

N«.I

o«.

IIIcIII.Ino: :.
IIIIIu I II.»
o. I:..IIII

 

oo.«

no.

«N.

.Cn.o

Inn.

IIhI.

IIIIIII.Ino:
n II.»

 

oo.«

N«.

o«.

IInI.

IIvn.

ICC“.

IIIcIII.Iaoz
I:..IIII

 

..00 I

hO.I

IONII

»«.

hu.I

ItILG
IIIu
I>.IuInIo

 

Io.I

»«.

»«.

IIoouao
IIIu
I>.III.I:I

 

IIu‘.

no.

IocIILIILIa
IIIu Lo.
Io.u:a.huu(

 

N«.

Ichv.

u:III>I. u<
go. I. an
.Ico.u:n.kuu<

 

N«.

II.IIIL0I0
c-.u<(x
I II.»

 

co.IIILan
augu<<x
N II.»

 

oo.«

co.IIILqu
auau<¢z
I:..III0

 

 

Isa-um
acI>u

0..”
I>.0I Oz

 

ALII
III>I~

III.

 

IIIcIII.Ino:

c. IucIsu
M II.» 0.
I:..III0

 

IIIcIII.
Iona:
n II.»

 

IIIcIII.
IIno:
I:..
IIIII

 

IIILG
IIxu
I) I

III 30

 

IIouuao
IIIu
II.»

-oInIII

 

Incl—Lo.
ILIa IIIu
.0. co.«
Ian.Luu<

 

IIIII>I.go(
Lo» I.hum
.Ico.u
Ian.Luu<

 

I‘a
¢I40(¢!
0 IE.»

 

I‘a
CIAU‘dl
N II.»

 

.nIo
coau¢¢z
I:..III0

 

 

 

 

2 0.0I»

152

 

LIST OF REFERENCES

Abramson, L. Y., Metalsky, G. l. & Alloy, L. 8., (1986). W
th f r in:Amtth rtiln iwithimlitinfr

h thol r r h. Manuscript submitted for publication.

 

Abramson, L. Y., Metalsky, G. I. & Alloy, L. 8., (1988). The hopelessness
theory of depression: Does the research test the theory? In L. Y.
Abramson (Ed.), i I n'i n
New York: Guilford Press.

Abramson, L. Y., Metalsky, G. I. & Alloy, L. 8., (1989). Hopelessness
depression: A theory-based subtype of depression. P h I i I
Beview, fi. 358-372.

Abramson, L. Y., Seligman, M. E. P., & Teasdale, J. D. (1978). Learned
helplessness in humans: Critique and reformulation. mm

mm: .81: 49-74-

Alloy, L. 8., Abramson, L. Y., Metalsky, G. I. & Hartlage, S. (1988). The
hopelessness theory of depression: Attributional aspects. _B_rjt_is_l3

rnl f lini I ,21,5-21.

Alloy, L. 8., Clements, C., & Kolden, G. (1985). The cognitive diathesis-
stress theories of depression: Therapeutic Implications. In S. Fleiss &

R- Bootzin (Eds). 3W9! 379-410. New

York: Academic Press.

Alloy, L. 8., Hartlage, S., & Abramson, L Y. (1988b). Testing the cognitive
diathesis-stress theories of depression: Issues of research design,
conceptualization, and assessment. In L. 8. Alloy (Ed), Qggnitive

W, 31-73. New York: Guilford.

153

Barnett, P. A., & Gotlib, l. H. (1988). Psychosocial functioning and depression:
Distinguishing among antecedents, concomitants, and consequences.

I i I ll in,M,97-126.

Beck, A. T. (1967).D r in: lini l x rim n l n th
New York: Harper & Row.

Beck, A. T. (1976). Cegnitive therepy end the emetienel eieereege. New York:

International Universities Press.

Beck, A. T., Ward, C. H., Mendelson, M., Mock, J., & Erbaugh, J. (1961). An

inventory for measuring depression. Ar hiv f n r IP hi t , 5,
53-63.

Beck, A. T., Weissman, A., Lester, D., & Trexler, L. (1974). The measurement
of pessimism: The hopelessness scale. m l f ltin Ii i

W. 52, 861-865.

Brewin, C. R. (1985). Depression and causal attributions: What is their
relation? Peyehelegieel Bulletin, 38, 297-309.

Cochrane, R., & Robertson, A. (1973). The life events inventory: A measure of
the relative severity of psychosocial stressors. mm

W. 12. 135-139

Cohen. J.. & Cohen. P. (1983)- W
W (2nd ed.). Hillsdale, NJ: Erlbaum
Coyne, J. C., & Gotlib, l. H. (1983). The role of cognition in

depression: A critical appraisal. P I i I II 'n, 3. 472-505.

Coyne, J. C., & Gotlib, I. H. (1986). Studying the role of cognition in depression:

Well-trodden paths and cuI-de-sacs. W,
_1_Q, 695-705.

Coyne, J. C., & Gotlib, I. H. (1983). Tne role of cognitions in depression: A
critical appraisal. h I i I ll ti , %, 472-505.

Crittenden, K. S., & Lamug, C. B. (1988). Causal attribution and depression: A
friendly refinement based on Philippine data. MW
Embolesm. .19. 216-231.

Crocker, J., Alloy, L. B., & Kayne, N. T. (1987). Depression, self-esteem, and
attributional style. Manuscript under editorial review.

 

 

 

154

Cutrona, C. E., Russell, D., & Jones, R. D. (1985). Cross-situational consistency
in causal attributions: Does attributional style exist? Jeernel ef Pereenelgy‘

n iIP hl ,4_7,1043-1058.

Depue, R. A. & Monroe, 8. M. (1978). Learned Helplessness in the perspective
of the depressive disorders. Jeernel ef Aenermel Peyehelegy, _81, 3-20.

Dweck, C. S., & Licht, B. G. (1980). Learned helplessness and intellectual
achievement. In J. Garber and M. E. P. Seligman (Eds), Humen

l l n :Th n Ii ti n. New York: Academic Press.

Hammen, C. L., Mayol, A., deMayo, H., & Marks, T. (1986). Initial symptom
levels and the life event-depression relationship. Jeergel ef Aenermel

Esmboicgy. es. 114-122.

Heimberg, R. G., Klosko, J. 3, Dodge, C. S., Shadick, H., Becker, R. E., &
Barlow, D. H. (1989). Anxiety disorders, depression, and attributional
style: A further test of the specificity of depressive attributions.

Qegnitive Therapy ene Reeeereh, 13, 21-36.

Hill, A. B., & Kemp-Wheeler, S. M. (1986). Personality, life events and

subclinical depression in students. Pereenelgy‘ ene Ineivieuel
Differeneee, 1, 469-478.

James, L. H., & Brett, J. M. (1984). Mediators, moderators, and tests for
mediation. m l fA Ii h l , a, 307-321.

Janoff-Bulman, R. (1979). Characterological versus behavioral self blame:

Inquiries into depression and rape. rn I fP r n l' n i l
W. .31. 1798-1809.

Kendall, P. C., Hollon, S. 0., Beck, A. T., Hammen, C. L., & Ingram, R. E.
(1987). Issues and recommendations regarding use of the Beck

Depression Inventory. W. Li. 289-299.

Lewinsohn, P. M., Steinmetz, J. L., Larson, D. W., & Franklin, J. (1981 ).
Depression related cognitions: Antecedent or consequence? Meiji

MW. 91. 213-224.

Metalsky, G. l., Abramson, L. Y., Seligman, M. E. P., Semmel, A., &
Peterson, C. R. (1982). Attributional styles and life events in the
classroom: Vulnerability and invulnerability to depressive mood reactions.

rnl fPr nl' n ilP l 43.612617.

155

Metalsky, G. I., Halberstadt, L. J., 8: Abramson, L. Y. (1987). Vulnerability
to depressive mood reactions: Toward a more powerful test of the
diathesis-stress and causal mediation components of the reformulated

theory of depression. rn l fP r n l' n i IP h I , 5g,
386-393.

Miller, I. W., Klee, S. H., & Norman, W. H. (1982). Depressed and
nondepressed inpatients’ cognitions of hypothetical events, experimental

tasks and stressful life events. rn I f n rm lP h l .91,
78-81.

Norman, P. D., & Antaki, C. (1988). Real events attributional style
questionnaire. rn l f i l n lini IP h I ,1, 97-100.

Nunnally, J. C. (1978). Eeyehemetrie theery (2nd ed.). New York: McGraw-Hill

Overmier, J. B., & Seligman, M. E. P. (1967). Effects of inescapable shock
upon subsequent escape and avoidance learning. Jeumajgj

Comparative end Phyeielegiml Peyehelegy, Q3, 22-33.

Peterson, C. H., & Seligman, M. E. P. (1984). Causal explanations as a risk

factor for depression: Theory and evidence. W, .91,
347-374

Peterson, 0., Semmel, A., von Baeyer, C., Abramson, L Y., Metalsky, G. l., 8.
Seligman, M. E. P. (1982). The attributional style questionnaire.

Qegnitive Therepy end Beeeeren, 5, 287-300.

Pittman, T. S., & Pittman, N. L (1980). Deprivation of control and the
attribution process. m I fP r n l' n i IP
377-389.

Pyszczynski, T. A. & Greenberg, J. (1981). Role of disconfirmed expectancies
in the instigation of attributional processing. W
W 59 3198

Riskind, J. H., & Rholes, W. S. (1984). Cognitive accessibility and the
capacity of cognitions to predict future depression: A theoretical note.

WW. 5. 1-12.

Rizley, R. (1978). Depression and distortion it the attribution of causality.
Wolsey. BL 3248.

I I

156

Segal, Z. V., & Shaw, B. F. (1986a). Cognition in depression: A reappraisal of

Coyne and Gotlib’s critique. W. 1.0.
671-693.

Segal, Z. V., & Shaw, 8. F. (1986b). When cul-de-sacs are more mentality than

reality: A rejoinder to Coyne and Gotlib. W
m: JD: 707'714-

Seligman, M. E. P. (1975). I I ' i n v I m nd

_deat_h. San Francisco: Freeman.

Seligman, M. E. P. (1978). Comment and Integration. Jeurpal ef Apnermal
mm: .81: 165'179-

Seligman, M. E. P., & Maier, S. F. (1967). Failure to escape traumatic shock.
Jeumal pf Experimental Peyehelegy. 13, 1-9.

W. (1988). (3rd ed.) Chicago, IL: SPSS Inc.

Sweeny, P. D., Anderson, K., & Bailey, S. (1986). Attributional style in

depression: A meta-analytic review. m I f r n I' n i l
W: m: 974-991-

Tennen, H. & Herzberger, S. (1987). Depression, self-esteem and the absence

of self-protective attributional biases. rn l fP r n l' n i l
Radiology. 52. 72-80.

Weiner, B. (1985). An attributional theory of achievement motivation and

emotion. W. 22. 548-573.

Williams, J. M. G. (1985). The attributional formulation of depression as a
diathesis-stress model: Metalsky et al. reconsidered.,1pumal_o_

W 53. 1572-1575.

Wong, P. T., & Weiner, B. (1981). When people ask “why“ questions, and the

heuristics of attributional search. I fP l' n i l
25mm. 59. 650-663.

Zuckerman, M., & Lubin, 8. (1965). f M ' If A ' iv
Queekjet. San Diego, CA: Educational and Industrial Testing Service.

Zuckerman M & Lubin. 3 (1985) WM
Weed San Diego CA: Educational and

Industrial Testing Service.

 

MICHIGAN STATE UNIV. LIeRnRIEs
illllWillilllllllWilliIllllllllllllllllllllllllllllllll
31293008989331