ms. 3'"

v1
t

.5”

 

THESlS

\
rfi

C; 07.74
éfl/Q7f—2 3’

This is to certify that the
dissertation entitled

MATERNAL ATTRIBUTIONS ABOUT CHILD DISRUPTIVE
BEHAVIORSIN RELATION TO CHILD BEHAVIORS AND
MATERNAL ADJUSTMENT

presented by

Cheryl-Lynn Podolski

has been accepted towards fulfillment
of the requirements for the

PhD. degree in Clinical Psycholoa

 

 

 

/ Major oféséofs Signature
§/§/Ay

August 3, 2004

MSU Is an Alfinnative Action/Equal Opportunity Institution

 

 

LIBRARY

Michigan State

University

 

 

PLACE IN RETURN Box to remove this checkout from your record.
To AVOID FINES return on or before date due.
MAY BE RECALLED with earlier due date if requested.

 

DATE DUE

DATE DUE

DATE DUE

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

6/01 cJClRC/DateDuepes-cts

__._——._

MATERNAL ATTRIBUTIONS ABOUT CHILD DISRUPTIVE BEHAVIORS
IN RELATION TO CHILD BEHAVIORS AND MATERNAL ADJUSTMENT

By

Cheryl-Lynn Podolski

A DISSERTATION

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

DOCTOR OF PHILOSOPHY
Department of Psychology

2004

ABSTRACT

MATERNAL ATTRIBUTIONS ABOUT CHILD DISRUPTIVE BEHAVIORS
IN RELATION TO CHILD BEHAVIORS AND MATERNAL ADJUSTMENT

By

Cheryl-Lynn Podolski

The pathways by which maternal attributions about child ADHD and other
disruptive behaviors relate to maternal adjustment are not well understood, yet such
understandings might contribute to effective interventions. Early studies indicated that
. maternal attributions about child problem behaviors influence both parental afi‘ective and
behavioral responses and subsequent behavioral chains. The current study applied this
line of research to child attention deficit hyperactivity disorder (ADHD), by investigating
maternal attributions about their children’s inattentive, overactive, and disobedient
behaviors in relation to maternal role adjustment.

Participants were 125 mothers for whom interviews were obtained about child
behaviors (DISC-IV), maternal attributions (Johnston’s Written Analogue Questionnaire
and Recalled Incident Interview), and maternal adjustment (Parenting Satisfaction Scale
and Parenting Stress Index). Child ADHD status was assessed via parent and teacher
ratings and the structured diagnostic interview. 51 children had one of the ADHD
subtypes, 34 were ADHD NOS/Subthreshold, and 41 were controls.

Similar to prior findings, mothers of children with ADHD characterized their
children’s misbehavior as more stable, global, and less controllable compared to mothers
of children without ADHD (F[1,81]=12.6, p<.001). This pattern is consistent with the

realities of ADHD. It is similar to the pattern of attributions associated with learned

helplessness but somewhat different from a pattern associated with distress in parents of
non-disordered children. In the current study, maternal attributions of stability (r=.34,
p<.001) and globality (r=.43, p<.001) were related to maternal role dissatisfaction and
role stress. Maternal attributions about child disobedient behaviors were related to
maternal adjustment even when ratings of child oppositional behaviors were controlled
(partial r=.25, p<.01). In contrast, the relation between attributions about child
inattentive/overactive behaviors and maternal adjustment was fully mediated by ratings
of child behaviors. Relatedly, child ADHD diagnosis moderated the relation between
maternal attributions about inattentive/overactive behaviors and maternal adjustment but
not the relation between maternal attributions about oppositional behaviors and maternal
adjustment.

In a secondary study, child attributions about their own behaviors were also
examined using a version of the Written Analogue Questionnaire adapted for use with
children. Children with ADHD attributed their disruptive behaviors to more
uncontrollable and internal causes compared to children without ADI-ID.

Findings are discussed in terms of implications of attribution theory for parental
adjustment to child ADHD and externalizing behavior, the possibility of unique processes
operating in families of children with ADHD versus non-ADHD families, the importance
of distinguishing symptoms of ADHD from oppositional behaviors, and potential
implications and suggestions for future intervention studies to evaluate the import of the
findings. Future work will examine paternal attributions. Overall, findings suggest that
parental and child attributions are an important element in family processes related to

development of childhood ADI-ID.

ACKNOWLEDGMENTS

With never-ending energy and dedication to quality, my mentor, Joel T. Nigg,
Ph.D., provided constructive analysis, guidance, and support invaluable to this work and
my development as a psychologist. Thank you for lending your acuity toward the
refinement of my work and own analytic mind. My co-chair, Alytia Levendosky, Ph.D.
provided me with encouragement and inspiration as I strove to achieve balance in my life
as well as excellence in my work. Thank you for your steadfast example and support.

Guidance and thoughtful analysis from my committee members, Judy Brady,
Ph.D., Robert Caldwell, Ph.D., and Rick Deshon, Ph.D., help me to sharpen my ideas and
consider multiple implications for my work. I would also like to thank Charlotte
Johnston, Ph.D., Wendy Freeman, Ph.D., and Besty Hoza, Ph.D., for allowing me to use
their measures and providing an exemplary precedent into attribution and family process
research in relation to ADHD.

The current study would not be possible without each of the families who
participated. I hope this research may be used to encourage and assist them. Many
others assisted me in this endeavor. My partner, Todd Bacon, was of endless support as
he worked with me through the trenches of formatting requirements, sacrificed hours
away from me as I sat at my computer, and shared the joys of the completion of this
project —— my never-ending thanks and appreciation to him for being there through the
process. The support of friends, colleagues, and family was also cherished along the way

and remains greatly valued.

iv

TABLE OF CONTENTS

LIST OF TABLES ............................................................................................................ IX
INTRODUCTION .............................................................................................................. 1
CHAPTER 1
LITERATURE REVIEW FOR STUDY 1: MATERNAL ATTRIBUTIONS ................... 2
Overview ......................................................................................................................... 2
Child Disruptive Behaviors ............................................................................................. 4
Parental Adjustment to Children’s ADHD ..................................................................... 5
Context and ADI-[D ......................................................................................................... 6
Attributions and Coping .................................................................................................. 9
Attributions, Adjustment, and Evaluations of Others ................................................... 10
Appraisal Versus Attribution. ................................................................................... 10
Weiner’s Attribution-Behavior Approach. ............................................................... 11
Attributions and Response to Others ........................................................................ 12
The “Learned Helplessness Model” of Attributions ................................................. 14
Dweck’s Theory-Attribution-Behavior Approach. ................................................... 16
Maternal Attributions .................................................................................................... 21
Maternal Attributions and Adjustment in Community Samples ............................... 21
Attributions in Parents of Children With Behavior Disorders. ................................. 23
A Self-Fulfilling Prophecy ........................................................................................ 25
Attributions made by Parents of Children with ADHD ............................................ 26
Attributions and Adjustment When a Child has ADHD ............................................... 29
Summary and Critique .................................................................................................. 30
CHAPTER 2
LITERATURE REVIEW FOR STUDY 2: CHILD ATTRIBUTIONS .......................... 32
“Helplessness”and “Mastery” Attributions ................................................................... 33
Performance Attributions by Children With ADHD. ................................................... 34
Helpless Style ............................................................................................................ 35
Protective Style ......................................................................................................... 37
Attributions About Disruptive Behaviors ..................................................................... 41
Critique of Current Findings with ADHD Children ................................................. 43
Medication Effects ........................................................................................................ 46
Conclusions from Literatures: Attributions and ADHD .............................................. 47
RATIONALE AND PLAN OF STUDY .......................................................................... 49
CHAPTER 3
METHOD ......................................................................................................................... 52
Participants .................................................................................................................... 52
Procedure ...................................................................................................................... 53
Measures ....................................................................................................................... 54
Child Behavior .......................................................................................................... 54

Inattention-hyperactivity. ...................................................................................... 54

Behavior Assessment System for Children ........................................................... 54
Conners’ Scales-Revised-Short Forms ................................................................. 55
Child behavior: Oppositional/disobedient. .......................................................... 56
ADHD Diagnosis. ................................................................................................. 56
Parent Attributions .................................................................................................... 58
Written Analogue Questionnaire. ......................................................................... 58
Recalled Incident Interview .................................................................................. 59

The Interactions Questionnaire ............................................................................. 60
Parent Role Adjustment ................................................................................................ 61
Satisfaction with Parenting Performance Scale ........................................................ 61
The Parenting Stress Index- Short Form ................................................................... 62
Measure of Child Attributions ...................................................................................... 62
Written analogue questionnaire. ............................................................................... 62
Additional Child Variables ........................................................................................... 63
Reading Disability (RD). .......................................................................................... 63
ODD or C D Diagnoses. ........................................................................................... 63
Data Reduction .............................................................................................................. 64
Data Reduction of Adult Reported Ratings of Child Behavior .................................... 64
Composites for Maternal Ratings of Child Behaviors. ............................................. 66
Teachers’ Ratings of Child Behaviors. ..................................................................... 67
Correlations Between Maternal and Teacher Data. .................................................. 68
Data Reduction of Maternal Attribution Data .............................................................. 68
Written Analogue Questionnaire (WAQ) ................................................................. 69
Creation of Subscales and Reliabilities ................................................................. 73
Recalled Incident Interview (RII) Data Reduction. .................................................. 75
Data Reduction for Measures of Maternal Role Adjustment ........................................ 81
Parent Satisfaction Survey and Parenting Stress Index Scales. ................................ 81
Additional Attribution Measures .................................................................................. 82
Attribution Items Pertaining to Parent Responsibility and Control .......................... 82
RH Attribution Items Pertaining to Parent Responsibility and Control .................... 84
Hoza Interactions Questionnaire ............................................................................... 85
Child Attribution Measures ........................................................................................... 88
Child Written Analogue Questionnaire .................................................................... 89
Child Ratings of Inattentive-Overactive Scenarios ............................................... 91
Child Ratings of Oppositional/Disobedient Scenarios ......................................... 92
Child Ratings of Positive/Prosocial Scenarios ...................................................... 93
Child HOZA .............................................................................................................. 95
Behavioral Compliance ......................................................................................... 96
Behavioral Non-Compliance ................................................................................ 98
Summary Regarding Child Attribution Measures ........................................................ 98
Primary Hypotheses .................................................................................................... 100
Hypothesis 1: Maternal attributions and child ADHD ........................................... 100
Hypothesis 2: Maternal attributions and role adjustment ....................................... 101
Hypothesis 3:Moderation by child diagnosis .......................................................... 102
Exploratory Analyses. ................................................................................................. 103

vi

Exploratory Analysis 1: Maternal Attributions and child dimensional behaviors. 103

Exploratory Analysis 2: Maternal responsibility and control ................................ 104
Exploratory Analysis 3: Child attributions ............................................................. 104
CHAPTER 4
RESULTS FOR STUDY 1 ............................................................................................. 106
Maternal Attributions .................................................................................................. 106
Hypothesis 1: Maternal attributions and child disruptive behaviors ..................... 106
Hypothesis la: Maternal Attributions and actual child diagnosis ..................... 106
Controlling for child aggressive behavior ....................................................... 108
Controlling for medication status on subset of sample ................................... 108
Subtype Analyses ............................................................................................ 108
Hypothesis 1b: Maternal Attributions and beliefs about diagnosis ................... 111
Additional analysis: maternal attributions and dimensional child behaviors 115
Hypotheses 2 and 3. Maternal Attributions and Role Adjustment ........................ 116
Hypothesis 2a: Dimensional relation between attributions and stress. .............. 117
Examination By Subtype ................................................................................ 118
Hypothesis 2b. Independence of maternal attributions. ..................................... 121
Additional analysis: Moderation by severity of child behavior .......................... 125
Hypothesis 3 — Moderation by Child Diagnosis ................................................. 128
Summary of Maternal Attribution Findings ........................................................... 130
Covariates ............................................................................................................... 133
Exploratory Analyses 1: Maternal Attributions and child behaviors ...................... 134
Exploratory Analysis 2: Maternal responsibility and control ................................. 136
Exploratory Analysis 3: Mediation by maternal attributions .................................. 141
CHAPTER 5
RESULTS FOR STUDY 2 ............................................................................................. 144
Child Results ............................................................................................................... 144
Exploratory Analysis 4: Child attributions ............................................................. 144
Exploratory Analysis 4a: Differences by diagnostic group ................................... 144
Exploratory Prediction 4b: Child attributions and adult ratings of behavior. ......... 145
Subtype Comparisons ......................................................................................... 146
CHAPTER 6
DISCUSSION ................................................................................................................. 149
Replication of Group Effects ...................................................................................... 151
Maternal Attributions and Belief About Child’s Diagnosis ....................................... 162
ADHD DSM-IV Subtype Differences in Attributions ................................................ 162
Maternal Attributions and Role Adjustment ............................................................... 164
Moderation of Attribution-Adjustrnent Relation by Diagnosis .................................. 167
Diagnostic versus dimensional considerations ........................................................... 170
ADHD Subtype Differences in Attribution - Adjustment Relation ............................ 171
Summary of Maternal Attribution Findings ............................................................... 171
Possible Implications Pertaining to Intervention Research ........................................ 172
Child Attributions ....................................................................................................... 174

vii

Limitations of Current Studies and Cautions when Interpreting Findings ................. 175

Conclusion .................................................................................................................. 177
APPENDICES ................................................................................................................ 180
REFERENCES ............................................................................................................... 190

viii

LIST OF TABLES

Table 1 Child sample characteristics by diagnostic group ................................................ 53
Table 2: Correlations between three-factor solution for mother ratings, (reliabilities on
diagonal) ........................................................................................................................... 66
Table 4 Correlations for teacher behavior rating composites, 3-factor solution
(reliabilities on diagonal) .................................................................................................. 68
Table 5 Correlations between mom and teacher ratings for three-factor solution ........... 68

Table 6 Factor loadings of Maternal Attributions for Inattentive /Overactive behaviors
via WAQ Questionnaire Scenarios ................................................................................... 70

Table 7 Factor Solution of Maternal Attributions for Oppositional / Disobedient
behaviors via WAQ Questionnaire Scenarios ................................................................... 71

Table 8 Final Factor Solution of Maternal Attributions for Oppositional / Disobedient
behaviors via WAQ Questionnaire Scenarios (globality items dropped) ......................... 72

Table 9 Factor loadings of Mom Attributions for Positive/Prosocial Behaviors via WAQ
Questionnaire Scenarios .................................................................................................... 73

Table 10 Intercorrelations among Mother rated WAQ factors and data reliabilities (on
diagonal) of composites .................................................................................................... 74

Table 11 Factor Solution of Maternal Attributions for Inattentive/Overactive behaviors
via RII Questionnaire Scenarios ....................................................................................... 76

Table 12 Factor Solution of Maternal Attributions for Oppositional/Disobedient
behaviors via RII Questionnaire Scenarios ....................................................................... 77

Table 13 Factor Solution of Maternal Attributions for Oppositional/Disobedient
behaviors via RII Questionnaire Scenarios (globality items dropped) ............................. 77

Table 14 Factor Solution of Maternal Attributions for Positive/Prosocial behaviors via
RII Questionnaire Scenarios ............................................................................................. 78

Table 15 Reliabilities of (on diagonal) and intercorrelations among mother rated RII
factor composites .............................................................................................................. 79

Table 16 Correlations between WAQ and RH subscales ................................................. 80

ix

Table 17 Reliabilities and inter-scale correlation matrix for maternal role adjustment .. 81

Table 18 Maternal WAQ Attributions about parent role in child behavior factors for
Inattentive-Overactive Behaviors ..................................................................................... 83

Table 19 Mother WAQ Attributions about parent role in child behavior factors for
Oppositional/Defiant child behaviors ............................................................................... 83

Table 20 Reliabilities for mothers’ WAQ attributions pertaining to her role in child’s
behavior, by behavior type ................................................................................................ 83

Table 21 Maternal RII Attributions about parent role in child behavior factors for
Inattentive-Overactive Behaviors ..................................................................................... 84

Table 22 Maternal RH Attributions about parent role in child behavior factors for

Oppositional/Defiant child behaviors ............................................................................... 84
Table 23 Reliabilities for mothers’ RII attributions pertaining to her role in child’s
behavior, by behavior type ................................................................................................ 85
Table 24 Factor structure for Hoza maternal attributions about own parenting and effort
........................................................................................................................................... 87
Table 25 Hoza four factor subscale reliabilities and inter-correlations. .......................... 88
Table 26 Factor analysis of child WAQ attribution ratings pertaining to Inattentive
Overactive child behaviors ............................................................................................... 91
Table 27 Factor analysis of child WAQ attribution ratings pertaining to Inattentive
Overactive child behaviors ............................................................................................... 92
Table 28 Factor analysis of child WAQ attribution ratings pertaining to
Oppositional/Disobedient child behaviors ........................................................................ 93
Tab1e29 Factor analysis of child WAQ attribution ratings pertaining to
Oppositional/Disobedient child behaviors ........................................................................ 93
Table 30 Factor analysis of child WAQ attribution ratings pertaining to
Positive/Prosocial child behaviors .................................................................................... 94

Table 31 Inter-correlations among child WAQ attribution ratings, data reliabilities on the
diagonal ............................................................................................................................. 95

Table 32 Factor analysis of child Hoza attribution ratings pertaining to Compliant
behaviors ........................................................................................................................... 96

Table 33 Three-factor solution of child Hoza attribution ratings pertaining to Compliant

behaviors ........................................................................................................................... 97
Table 34 Two-factor solution for child Hoza compliance items ..................................... 97
Table 35 Correlations between child WAQ and Hoza ratings ......................................... 99

Table 36 Test of differences in attributions made by mothers of children with ADHD
diagnosis versus mothers of children without a behavioral disorder, using the WAQ
measure that describes child behavior problems. ........................................................... 107

Table 37 Results of maternal attributions about child behavior, testing for group
differences when comparing the independent variable of child diagnosis with four levels
(ADHD-C, ADD, subthreshold, and controls), dependent variable=attribution domain by
behavior type ................................................................................................................... 1 10

Table 38 Cross-tabulation of child diagnosis with mothers’ belief about diagnosis ..... 112

Table 39 Test of differences in attributions made by mothers of children who indicated
that they “believe their child has ADHD” (whether or not our testing confirmed or
discontinued this) versus mothers who believe their child “does not have ADHD);
attributions as measured by the WAQ standardized scenarios in which to imagine own
child engaging. ................................................................................................................ 114

Table 40: Correlations between maternal attributions and child behaviors ................... 115

Table 41 Correlations between maternal attributions (using scenario measure, WAQ) and
parent role stress and satisfaction (2-tai1ed) .................................................................... 117

Table 42 Correlations between maternal attributions (using scenario measure, WAQ) and
parent role stress and satisfaction (l-tailed) .................................................................... 117

Table 43 Correlations between maternal attributions (using scenario measure, WAQ) and
parent role stress and satisfaction (2-tailed) .................................................................... 119

Table 44 Correlations between maternal attributions (using scenario measure, WAQ) and
parent role stress and satisfaction (l-tailed) .................................................................... 120

Table 45 Regression to test relation between maternal attributions and maternal role
adjustment independent of child behaviors (step 3) and test of moderation (step 4); with
dependent variable=PSS Satisfaction with Parenting for all models .............................. 122

Table 46 Regression to test relation between maternal attributions and maternal role

adjustment independent of child behaviors (step 3) and test of moderation (step 4); with
dependent variable=PSI SF Total Stress for all models ................................................. 123

xi

Table 47 Correlations between maternal attributions about child behavior in relation to
mother and teacher ratings of child behaviors (2-tai1ed) ................................................ 125

Table 48 Regression to test relation between maternal attributions and maternal role
adjustment independent of child behaviors (step 2) and test of moderation (step 3); with
dependent variable=PSS Satisfaction with parenting performance (controlling only for
behavior related to the attribution of interest) ................................................................. 127

Table 49 Regression to test relation between maternal attributions and maternal role
adjustment independent of child behaviors (step 2) and test of moderation (step 3); with
dependent variable=PSI-SF Total Stress (controlling only for behavior related to the
attribution of interest) ..................................................................................................... 128

Table 50 Regression to test moderation of diagnostic variable (control vs. any type of
ADHD) in the relation of maternal attributions and role adjustment, dependent
variable=PSS satisfaction with parenting performance .................................................. 129

Table 51: Regression to test moderation of diagnostic variable (control vs. any type of
ADHD) in the relation of maternal attributions and role adj ustrnent, dependent
variable=PSI Total Stress as outcome ............................................................................ 130

Table 52 Summary of primary findings ......................................................................... 132

Table 53 Correlations between maternal attributions (using scenario measure, WAQ) and
parent role stress and satisfaction, with child aggression/oppositional behaviors controlled
(2-tailed) .......................................................................................................................... 133

Table 54 Correlations between maternal attributions (using scenario measure, WAQ) and
parent role stress and satisfaction, with child aggression/oppositional behaviors controlled
(1-tailed) .......................................................................................................................... 134

Table 55 Correlations between maternal attributions about child behavior in relation to
mother and teacher ratings of child behaviors (3-factor analysis) (2-tailcd) ................. 136

Table 56 Correlations between attributions about parent’s own role in child behavior
(attributions of responsibility and control) in relation to parents’ role adjustment (PSS
satisfaction and PSI Total Stress). (2-tailed) ................................................................... 138

Table 57 Moderation test controlling for child behavior, using RH; PSS as outcome
variable ............................................................................................................................ 140

Table 58 Correlations between child behaviors and maternal role adjustment (2-tailed)
......................................................................................................................................... 141

Table 59 Regression to test mediation of child behavior and maternal role adjustment by
maternal attributions with dependent variable=PSI-SF Total Stress .............................. 143

xii

Table 60 Test of differences in attributions made by children with ADHD diagnosis
versus children without a behavioral disorder, using the WAQ and Hoza measures 144

Table 61 Correlations between child attributions about child behavior in relation to
mother and teacher ratings of child behaviors (3-factor analysis) (2-tailed) .................. 145

Table 62: Correlations between child attributions about child behavior in relation to
mother and teacher ratings of child behaviors (3-factor analysis) (1-tai1ed) ................. 146

Table 63 Results of child attributions, testing for group differences when comparing the
independent variable of child diagnosis with four levels (ADHD-C, ADD, subthreshold,
and controls), dependent variable=attribution domain by behavior type ....................... 147

Table 64 Correlations between mother and child attributions about child behaviors... 148
Table 65 Studies pertaining to attributions made by children ....................................... 181
Table 66 Studies pertaining to attributions made by parents ......................................... 183

Table 67 Test of differences in attributions made by mothers of children with ADHD
diagnosis versus mothers of children without a behavioral disorder, using the RI] measure
that describes child behavior problems. .......................................................................... 185

Table 68 Results of maternal attributions about child behavior, testing for group
differences when comparing the independent variable of child diagnosis with four levels
(ADHD-C, ADD, subthreshold, and controls), dependent variable=attribution domain by
behavior type. Using Recalled Incident Interview to measure attributions ................... 186

Table 69 Correlations between maternal attributions (using scenario measure, R11) and
parent role stress and satisfaction ................................................................................... 187

Table 70 Results with R11. Correlations between maternal attributions about child

behavior in relation to mother and teacher ratings of child behaviors (3-factor analysis)
......................................................................................................................................... 1 88

Table 71 Regression to test mediation of child behavior and maternal role adjustment by

maternal attributions with dependent variable=PSI-SF Total Stress, controlling only for
child behavior corresponding to attribution) ................................................................... 189

xiii

INTRODUCTION

Surprisingly little is known about contextual factors related to behaviors
associated with Attention Deficit Hyperactivity Disorder (ADHD). Despite the relative
lack of attention to contextual factors, recent studies have repeatedly found that parents of
children with ADHD experience higher levels of stress than their contemporaries
(Anastopoulos, Guevremont, Shelton, & DuPaul, 1992; Johnston & Mash, 2001;
Podolski & Nigg, 2001). What factors contribute to this stress and what might be done
to ameliorate it are just beginning to be explored. Notably, parent adjustment to their
child’s ADHD is important to study in order to help parents help children. The current
study focused on one factor that may be important for understanding mechanisms that
contribute to maternal stress. Through understanding such mechanisms, it is possible that
better interventions for parents and their children may be designed.

A number of separate literatures indicate that cognitions are important in
determining an individual’s responses to others’ behaviors. These literatures include:
coping and adjustment, social cognition, and parenting and child development. These
literatures will each be briefly reviewed. Then following from and expanding upon these
literatures, parent attributions about child behaviors are proposed as an important factor

likely to influence parent and child adjustment.

CHAPTER 1
LITERATURE REVIEW FOR STUDY 1: MATERNAL ATTRIBUTIONS

Overview

In examining parent attributions as an important contextual factor relevant to
parental adjustment to their child’s ADI-II), a basic argument is made. First, child ADHD
behaviors are proposed as a stressor for parents. A body of literature supports this
assertion. The bi-directional pathways by which child behavior and parent stress
exacerbate each other is acknowledge and in fact is a support for the need for studying
parental adjustment. With this acknowledged, child ADHD behaviors are outlined as a
stressor for parents.

A second line of argument taken herein is that studying parent attributions and
adjustment is important not only in terms of helping parents but also in terms of helping
children. Possible implications for child adjustment are outlined in relation to parent
attributions and adjustment to clarify the importance of the study in terms of helping
children as well as parents.

Given the importance and need to study parent attributions, findings from general
coping literature, social cognitive literature, and recent studies of parent attributions are
reviewed. In brief, the learned helplessness model of attributions is reviewed, suggesting
that a pessimistic attributional style for negative events is associated with poor
adjustment. The domains of globality, stability, intemality, and uncontrollability for
negative events have been associated with depression (Peterson & Bunce, 1997;

Sweeney, Anderson, & Bailey, 1986). The literature on social cognition, affect, and

behavior illuminates attributions that might be made about child disruptive behaviors and
how these behaviors might in turn be linked to parental role adjustment. This literature
indicates that attributions about others effect affective or behavioral responses to those
others (W einer, 1980). Then the literature on parent attributions generally is reviewed.
This literature suggests that the ways parents perceive their children’s behavior likely
contributes or mitigates the stress they experience (Dix & Grusec, 1985; Podolski &
Nigg, 2001).

The few initial studies that have been conducted with mothers of children with
ADHD are then reviewed for comparison to the findings from the general literature.
Recent studies have found that parents of children with ADHD believe inattentive-
overactive and oppositional defiant behaviors are uncontrollable by the child and stable
across time and situation (Johnston & Freeman, 1997). It is noted that these attributions
coincide with what we might expect given that by diagnostic criteria children with
ADHD exhibit hyperactivity and inattentive behaviors in multiple settings and to levels
of severity such that the behaviors are not controllable by normal means. Thus, recent
findings coincide with the diagnostic criteria.

Notably, despite these findings, the ways in which these cognitions affect parental
affective and behavioral responses to their children are yet to be well understood. To
date, only one known study investigated attributions in relation to parent adjustment to
their child’s ADI-ID (Hoza et al., 2000). Thus, the current study aimed to replicate earlier
findings regarding the attribution profile parents of children with ADHD make about
their children’s behaviors and secondly to explore whether and in what ways these

attributions in turn relate to parental role adjustment.

Child Disruptive Behaviors

Child disruptive disorders serve as a documented source of stress for parents
(Anastopolous et al., 1992; Barkely, 1998; Podolski & Nigg, 2001). Disruptive behavior
disorders include Attention Deficit Hyperactivity Disorder (ADHD), Oppositional
Defiant Disorder (ODD), and Conduct Disorder (CD). Although ODD and CD has long
been understood as related to parent and family stress, ADI-ID has only recently been
understood as a source of stress for parents. Before reviewing that research, a brief
overview of ADHD is provided.

Attention-Deficit Hyperactivity Disorder (ADHD) is currently one of the most
prevalent childhood psychiatric disorders. Estimates suggest that 3 to 5% of school-aged
children exhibit severe enough levels of inattention and/or hyperactivity across multiple
settings such that the meet diagnostic criteria for ADHD (American Psychiatric
Association, 1994; Szatrnari, Offord, & Boyle, 1989a). ADHD is characterized by levels
of activity, irnpulsivity, and/or inattention that are extreme for developmental level and
severe enough to interfere with the child’s adjustment across settings. By diagnostic
criteria, the behavior of children with ADHD is such that the child’s behavior is not
readily controllable through normal parenting behaviors.

The child’s disruptive behaviors are also severe enough to interfere with
relationships and academic functioning. For example, children with ADHD are often
unable to pay attention to detail, organize, remember, or keep track of things (American
Psychiatric Association, 1994). These problems have real world implications in terms of

impaired academics, social and emotional problems, and stressful parent-child

interactions. Unsurprisingly, child ADHD behaviors are associated with parental role
stress (Flick, 1994; Johnston & Mash, 2001; Podolski & Nigg, 2001).

Aggressive and oppositional behaviors are the most common and serious co-
existing problems for children who exhibit inattentive/overactive behaviors. Co-
occurring aggression reaches diagnostic levels in 30-50% of ADHD cases (Anderson,
Williams, McGee & Silva, 1987). Some studies suggest that 54-67% of children and
adolescents with ADHD meet full criteria for Oppositional Defiant Disorder (ODD) or
Conduct Disorder (CD) (Barkley & Biederman, 1997; Faraone & Biederman, 1997).
Understandably, co-occurring aggressive behavior provides another source of stress for
parents of children with ADHD. It is a serious complication with implications for
children with ADHD and for their parents - with both disorders contributing uniquely to

parent stress.

Parental Adjustment to Children’s ADI-ID

Studies consistently have found that mothers of children with ADHD experience
higher levels of distress compared to parents of non-disordered children. This distress is
not limited to global psychological distress but refers to role specific parenting stress as
well. That is, mothers of children with ADHD reported greater global psychological
distress (Befera & Barkley, 1984), greater role specific stress (Mash & Johnston, 1983a),
and lowered sense of parenting competence (Mash & Johnston, 1983a) compared to
mothers of children without behavioral disorders. Severity of child ADHD symptoms
has been related to maternal parenting stress (Anastopolous et al., 1992; Podolski &

Nigg, 2001).

This role stress appears to be related to ADHD symptomatology as well as to co-
morbid aggression. Although early studies did not control for the effect of comorbid
aggression on maternal stress, recent studies indicate that children’s ADHD behaviors
contribute significantly to parent role stress independent of co-morbid aggression
(Anastopoulos et al., 1992; Johnston, 1996b; Podolski & Nigg, 2001). Anastopoulos et
a1. (1992) found that child CBCL Aggression (Achenbach, 1991) accounted for 37% of
the variance in parental stress and that CBCL Attention Problems accounted for a further
4% of variance (p<.001). Podolski and Nigg (2001) found that child inattention and
oppositional-conduct problems but not hyperactivity were uniquely related to role stress
in mothers but not fathers. These studies demonstrate that parents may experience
significant role stress specifically related to their child’s ADI-I'D behaviors, even though
much of their stress is in response to associated aggressive behaviors. Thus, child ADHD
as well as child aggression serve as significant and important stressors for parents
although the co-morbid aggression proves to account for a large but not exclusive portion

of the variance predicting parent role stress.

Context and ADHD
ADHD is currently viewed as a largely biogenetic behavioral disorder
(Biederman et al., 1992; Tannock, 1998); however, family and contextual mechanisms
are important to the development, maintenance, and exacerbation of the disorder
(Biederman et al., 1995; see review by Johnston & Mash, 2001) and in the effective

treatment of ADHD. For example, family contextual and parental factors likely impact

treatment decision-making and treatment adherence (Gage & Wilson, 2000;
Christophersen & Mortweet, 2001).

Notably, contextual factors appear related to the maintenance and exacerbation of
ADHD symptoms in children and to the development of comorbid aggressive behaviors
(Hinsbaw et al., 1997). The negative interactions that ensue between ADHD children and
their parents, teachers, and peers create stress and conflict between the child and others in
their environment and likely leads to frustration and increased stress. Increased stress in
parents is associated with ineffective parenting practices (Patterson, 1996). Although
medication is the primary treatment for ADHD, behavioral programs implemented by
parents and teachers are also important aspects of treatment. Increased parent stress and
related ineffective parenting strategies are precisely opposite of what is needed to help
children with ADHD. Notably, treatments which incorporate family contextual factors
appear to provide needed support to families whose children have ADHD (Barkley,
1990), decreasing disruptive child behaviors, addressing comorbid oppositional and
disobedient behaviors, and parent stress. Cognitive behavioral interventions have also
been successful in ameliorating child behavioral dysfunction and are often used in
conjunction with medication (and without medication in some situations) (Pisterman,
McGrath, Firestone, & Goodman, 1988; Pollard, Ward, & Barkley, 1983). The cognitive
behavioral and psychosocial interventions ameliorate child behaviors (Pisterman,
McGrath, Firestone, & Goodman, 1988; Pollard, Ward, & Barkley, 1983) and also result
in higher satisfaction and lowered parenting stress (compared to treatment of ADHD with
medication alone) (MTA Cooperative Group, 1999; Barkley, 1990). While it is

acknowledged, that ADHD symptoms are dramatically reduced by medication,

psychosocial and cognitive behavioral interventions clearly add to the treatment and
appear to further facilitate reduction of child problematic behaviors and to significantly
reduce parent stress. Additionally, cognitive behavioral interventions that incorporate
contextual factors likely contribute to the prevention and/or treatment of comorbid
oppositional/aggressive behaviors.

Notably, co-occurring aggressive behaviors in children with ADHD is a primary
predictor of poor long-term adjustment in children (Hechtrnan & Weiss, 1983; Hechtrnan,
Weiss, Perlman, & Amsel, 1984). How to prevent the co—occurrence of such problematic
behaviors is thus essential for helping parents and their children. The reasons for the
comorbidity of ADHD and aggression are multiple. Although bio genetic factors are
likely key as the child’s inability to suppress impulses and engage in planful behavior
likely contributes to aggressive acting out, the lack of environmental supports and
negative interactions with parents and other authorities is also a likely primary
determinant of the development of co-morbid aggression. Currently, it is believed that
although a common genetic vulnerability may contribute to the link between ADHD and
aggression (Plomin et al., 1990), variability of comorbidity rates over time supports the
role of environmental factors as a possible greater contributor (Frick, 1994; Lahey &
Loeber, 1998). The known causal influence of family stress and conflict in the
development of aggressive behavior in children means that children with ADHD are at
risk for the development of co-morbid aggression not only due to child’s irnpulsivity but
also because the family is at-risk via higher stress and strain. Thus, understanding
maternal adjustment to a child’s ADHD is important not only to prevent possible

exacerbation of ADHD symptomatology but also to prevent the development of

comorbid aggressive and oppositional/defiant behaviors. One possible causal link
between ADHD and aggression may be contextual factors such as maternal stress and the
attributions that affect adjustment.

Notably, maternal attributions have been linked to treatment decisions (Wright ct
al., 2000) and to treatment outcomes (Hoza, et al., 2000). For example, Hoza and
colleagues (2000) recently found that in parents of children with ADHD, parental
attributions about their children’s behaviors were (a) related to parenting behaviors and
(b) predictive of child treatment outcomes. Although not directly tested, the authors
hypothesized that parental emotional adj ustment may be a primary mediator between
parent attributions, parenting behaviors, and child treatment outcomes. The current study
examined parent attributions about their children’s misbehaviors as a link to parental

emotional adj ustment (as measured by role adjustment).

Attributions and Coping

Given the importance of maternal adjustment in relation to child adjustment and
the increasingly large literature on parent stress and ADHD, it is somewhat surprising
that so few studies have examined possible mediating factors between the link between
parent stress and their children’s ADHD. As will be discussed in more detail later, the
few studies that have examined factors that might ameliorate this stress (Anastopoulos et
al., 1993; Podolski & Nigg, 2001) suggest that maternal attributions about children’s
behavior are likely to be one important variable related to parental adjustment. Indeed,
literature in parents and normal adults suggests that attributions are determinant of

adjustment to stressful circumstances generally.

A large body of literature links attributions to an individual’s adjustment to
stressors. In fact, how one appraises a situation or particular behaviors is directly linked
to whether or not that situation/behavior is experienced as stressor (Lazarus & Folkman,
1984). Specifically, the interpretation one gives stressful events facilitates or impedes
adjustment to a wide range of stressors (Lazarus & Folkman, 1984). For example,
Lazarus and Folkrnan’s (1984) transactional theory of coping to stressfirl events is based
in part on the importance of appraisal in affecting emotional and behavioral responses to
stressors. According to their theory, appraisal of “whether one has personal stake” in an
event is directly related to one’s emotional response to an event (p. 272). Most theories
of coping assert that appraisal is significant in determining whether or not an event is
experienced as stressful (Lazarus & Folkman, 1984; McCubbin & Patterson, 1983) and as

an important mediator of behavioral response (Thompson et al., 1992).

Attributions, Adjustment, and Evaluations of Others
Appraisal Versus Attribution.

Appraisal refers to the evaluation of a situation or event for importance and
personal relevance (Lazarus & Folkman, 1984) whereas attributions generally refer to
beliefs or explanations about causes of the event (Miller, 1995). Just as appraisals of
events have significance for adjustment, a person’s explanations and perceptions of the
causes of events and of his/her agency in relation to that event are related to adjustment
(Dweck & Leggett, 1988; Peterson & Vaidya, 2001). A person’s attributions about the
causes of events have been viewed as determinant of his/her adjustment to those events.

Just as appraisal about whether or not an event is considered stressful and relevant to an

10

individual’s adjustment, a number of factors related to the cause of the event have also
been seen as determinant of the individual’s adjustment.

Two primary literatures on attributions are considered relevant to the current
study. First is a review of the social cognitive literature that focuses on how causal
attributions relate to evaluations of others. Second is a review of the ways in which

causal attributions relate to ones’ own adjustment.

Weiner ’s Attribution-Behavior Approach.

Initially, examination of parent attributions grew out of social cognitive studies of
adults (see review by Dix & Grusec, 1985). Studies of adults’ beliefs were examined in
relation to affective reactions and behavioral responses to others (W einer, 1980). A
number of theories were developed which stated that attributions held by one person
affected their response to the object of the attributions (see review by Fiske & Taylor,
1991). These theories were tested with research studies in order to find what types of
attributions were made under what conditions and whether particular attributional profiles
were related to differential responses by an observer and under what circumstances.

Weiner (1980) has been credited for explicating the three domains of causal
attributions (see review by Miller, 1995; see review by Dweck & Leggett, 1988). These
three domains are: locus (internal vs. external), stability (stable versus unstable), and
controllability (controllable versus uncontrollable). Intemality refers to explanations of
causes that are due to characteristics of the self rather to some external or contextual
cause. Stability refers to explanations or beliefs that the cause is consistent over time

rather than operate in a temporary fashion. Globality refers to explanations or beliefs that

11

the cause is “going to undermine everything” rather than pertain to a limited, specific
situation.

In addition to outlining these attributional domains, Weiner (W einer, Graham, &
Chandler, 1982; Weiner, Russell, & Lerman, 1978) is credited as one of the first
psychologists to extend attribution theory from analyses of behavior to include the
understanding of emotional reactions to events or people (see review by Miller, 1995).
Weiner (1980) argued for a “cognition (attribution) — emotion — action temporal
sequence” by which the causal attributions about an event preceded an emotional
response to that event that in turn affected in a determining way the behavioral response.
Weiner postulated that causal attributions are central to the affective adjustment as well
as behavioral responses to a stressor.

Weiner (W einer, Graham, & Chandler, 1982) found empirical support for the
cognition — emotion — action temporal sequence first in studies of academic achievement
(W einer, Russell, & Lerman, 1978) and then in terms of helping behaviors (W einer,

1980). Most relevant to the current study is Weiner’s work related to help giving.

Attributions and Response to Others
Although Weiner’s studies did not examine parenting attributions, he found
causal attributions to be related to one person’s emotional reactions to another. Help-
giving behaviors and positive, empathic emotional responses were related to attributions
that the cause of need was due to disability or illness (which might assumed to be
external) and was uncontrollable. The controllability dimension was seen to be of

particular importance. Additionally, Weiner concluded that the relations found between

12

attributions and behaviors “exists because meptions of control are strongly related to
affective reactions” (1980, p. 196). This work suggests that understanding parent
attributions about their children’s misbehavior may illuminate understanding of processes
‘ related to their emotional adj ustment to those behaviors. Additionally, Weiner’s work
allows the speculation that for parents external and uncontrollable attributions about child
misbehaviors would be related to empathy and pgssibly a more pgsitive, less stressed
reaction to their children’s behavioral problems. In his study of helping behaviors,
Weiner found that ratings of help-giving were lowest when the helper attributed the cause
of the need to factors internal to the person in need and when the need for help was
perceived as being within the person’s control.

In support of Weiner’s theory that attributions by care-providers are related to
affective and behavioral responses to persons in need, Dagnan, Trower, and Smith (1998)
found that attributions of controllability were reljated to negative emotion. In a study of
care staff workers’ responses to adults with severe learning disabilities, Dagnan, Trower,
and Smith (1998) found controllable (but uncontrolled) attributions by staff about patient

behaviors were associated with negative emotional responses to that behavior.

Controllability and negative emotion were associated with less willingness to help.

 

 

Extrapolating to an ADHD population, these findings may imply that attributions of
controllability made by parents about child misbehavior might be associated with more
negative emotion toward the child and more parental role stress (or inversely that
attributions of uncontrollability might be associated with less role stress).

Weiner also suggested that stability of behavior was an important attribution to be

considered. Notably, more than the other attribution domains, stability is directly related

13

to schema theory. Theories of attributions essentially postulate that individuals hold
schemas that affect resultant reactions and behaviors. Thus, Weiner’s theory is a social
information-processing model by which a person’s beliefs affect his/her emotional and
affective reactions. In terms of stability, the more stable a behavior is seen to be, the
more that belief or schema is placed onto additional situations. Thus, if a behavior is
seen as stable, then the set of attributions about that behavior may have more influence
than if the behavior were not seen to cut across situations. In this way, stability
attributions may be of particular importance in determining parental adjustment to their

child’s ADHD if those ADHD behaviors are seen as stable.

The “Learned Hehrlessness Mode ” of Attributions

Another line of research involving causal attributions is of potential relevance to
the current study. The three causal attribution domains were also found to be linked to a
phenomenon identified as “leamed helplessness” (see review by Abramson, Seligman, &
Teasdale, 1978 or Peterson, Maier, & Seligrnan, 1993).

As is widely taught in psychology textbooks, the concept of learned helplessness
originated in studies by which dogs that were shocked eventually abandoned escape
efforts following a series of conditions in which efforts to escape continuously met with
no result. The dogs were thought to have learned the independence between their actions
and what followed (response-outcome independence). After repeated exposures, the dogs
failed to try to escape the shock even when the barriers that had previously prevented

their escape were taken away.

14

The examination of people’s attributions following uncontrollable events led to
the extension of “learned helplessness” to people (Abramson, Seligrnan, & Teasdale,
1978). An analogy was drawn between the dogs “learned helpless” behaviors and
behaviors exhibited by individuals who exhibited depression. Through a series of
studies, it was discovered that stable, internal, and global attributions for uncontrollable,
negative events were associated with “helpless” or quitting behaviors (Peterson &
Barrett, 1987; Peterson, Maier, & Seligrnan, 1993) and depression (Brewin, 1985;
Peterson, etal., 1982; Peterson, Maier, & Seligman, 1993). Of note for comparison to
current study findings, Peterson has recently noted that stability and globality may
account for most of relation between attributional style and depression (Peterson &
Bunce, 1997; Peterson, Seligman, Yurko, Martin & Friedman, 1998).

Given that mothers of children with ADHD are faced with a disorder that by
definition is “uncontrollable by normal means,” stable, and global across situations, it is
possible that mothers would experienced depressed symptomatology in the face of their
child’s ADI-ID. Although beyond the scope of the current study, “helpless” or giving up
of particular parenting behaviors would also be expected. It is possible that ADHD
behaviors would also be attributed as internal to the child. Notably, while borrowing
from the learned helplessness literature in attempting to understand maternal stress, it is
important to note that within the learned helplessness literature, attributions about the
individual ’s own behavior were assessed. The current application is to ascertain whether
attributions about the child’s behavior (e. g., is the child’s behavior uncontrollable for the
child not for the parent). This is noted as an important difference between the leamed

helplessness literature and the current study measures.

15

While the attribution approach is seen as most useful for an initial examination of
the role of mothers’ beliefs and adjustment in relation to their children’s behaviors, a
third line of research which extends beyond attributions is worth mention. Dweck
(Dweck & Leggett, 1988) considered underlying theories that precede judgments and
reactions based on attributions about others behaviors. That is, Dweck has argued that
people have implicit theories fiom which their attributions stem. While beyond the scope
of the current study, it is worthwhile to make brief mention of this theory for future
consideration and for the understanding of further possible implications of the current
study findings. Additionally, Dweck’s research underscores which attribution domains

may be most relevant for study.

Dweck ’s Theory-A ttribution-Behavior Approach.

Dweck’s theory posits that people hold implicit theories about themselves and
others. Their goals and evaluations both of themselves and others are based on these
implicit theories. Thus, Dweck’s theory attempts to understand theories that precede
attribution evaluation. According to Dweck, individuals make attributions and set goals
based on implicit theories. Their responses (emotional and behavioral) to those
attributions and goals are thus first guided by the individual’s implicit theories. That is,
like Weiner, Dweck proposes an attribution-motivation-behavior sequence. Dweck’s
sequence is stated as theory -) goal orientation/attribution evaluation 9 behavior pattern
(Dweck & Leggett, 1988).

Dweck’s work has contrasted two primary theories about traits - traits as a fixed

entity versus traits as incremental and malleable (Dweck & Leggett, 1988). An entity

16

theory is “the belief that traits are fixed” versus incremental trait beliefs that “traits are
malleable” (Heyman & Dweck, 1998, p.391). A fixed/entity theory of behavior (and
motivation) indicates stability of the behavior while a malleable theory of behavior
indicates lack of stability of behavior. Similarly, with a fixed entity theory, one would
anticipate potentially more generalizability and less specificity.

According to Dweck (Heyman & Dweck, 1998), persons who ascribe to an entity
view of behavior are more likely to make helpless attributions. That is, if one believes
that a behavior or trait is fixed and thus cannot be changed, that person is more likely to
acquiesce to difficulties rather than believe that change is possible. In contrast, the
incremental or malleable view of traits suggests that change is possible and thus one with
this view may demonstrate more resilience in the face of difficulties. To the extent that
ADHD behaviors are viewed as unchangeable, stable, and global (all attributions
consistent with diagnostic criteria), one might expect more helpless or frustrated
behaviors. Notably, another noteworthy point about Dweck’s work is that even if one
ascribes to an entity view (i.e., attributes are fixed), control over events is still possible
(see p. 269, Table 6 of Dweck & Leggett, 1988). Whether or not, control over an event is
possible is seen to depend on the individual’s perception of the level to which they have
the attribute in question.

Dweck (Dweck & Leggett, 1988) uses ability and intelligence as examples. For
example, if the entity theory is ascribed for intelligence, whether or not one has control
over events depends on the level of intelligence. If intelligence is high, control over an
event (such as grade on a test) is possible (but not guaranteed). If intelligence is low,

control is viewed as not possible. Thus, control over an event depends on the

17

characteristics and level of the (unchangeable) trait. For incremental theorists, control is
seen as possible over the trait (intelligence in this example) and over events. However, if
the trait is low, then control will require more time and effort. The potential application
to the current study would be if ADHD is ascribed as an entity or incremental attribute.
Whether or not behavior related to ADHD is controllable would further depend on the
level of the attribute in the individual (this coincides potentially with whether or not an
individual is seen as meeting severity sufficient to warrant a diagnosis).

The goal orientation is also seen as important (although less relevant to the
current study). Using the example of intelligence, an entity theory of intelligence may
increase the likelihood of performance (rather than learning) goals. Consequently, failure
and struggle may result in helpless behaviors. In contrast, an incremental theory suggests
that personality/intelligence is malleable. The goal thus is to learn behaviors so as to
increase competence and the behavior is mastery oriented. Notably, Dweck’s research
focuses on implicit theories and the social inferences that stem from those theories.

While Dweck’s work originally focused on intra-individual factors, it has more
recently been expanded to consider judgments about others as well (Heyman & Dweck,
1998; Levy & Dweck, 1998). Recent research by Dweck and colleagues suggests that an
“entigl views of personality (and behavior) are associated with an increased tendency to
make global negative judments of others who commit transgressions (Levy & Dweck,
1998). Applying this to ADHD, an entity view might coincide with more negative views
of child behavior qualified by the quantity of the trait. In contrast to entity view, if the
trait is considered incremental or malleable, than judgments about that trait in oneself and

others are seen to be less focused on the quantity but to revolve more around factors as to

18

how the trait is displayed in that particular situation and at that particular point in time.
In essence, entity theory lends toward more stable and global attributions whereas the
incremental theory lends toward attributions that the trait or behavior is malleable, more
situation specific (less global) and less stable.

How might Dweck’s theory increase our understanding of maternal attributions
about their children’s ADHD behaviors? If ADHD is a fixed entity, control might be
possible if the degree of ADI-II) is low (which might be subthreshold ADHD symptoms).
If ADHD is viewed as an incremental trait, and thus malleable, control would be seen to
depend on time and effort. Current theories of ADI-1D, do not coincide entirely with
either an entity or incremental theory but rather ADHD as conceptualized today appears
to have elements of both. Before understanding the implicit theories about ADHD, first
understanding the particular attributes seems warranted, noting that a child’s ability to
control his/her behavior may be conceptualized both as a trait and an event. Adding in
the component about behavior and goals, Dweck’s work might suggest that parents who
ascribe to entity theory of ADHD (or attributions coinciding with entity theory) might be
less oriented toward learning goals or toward instructing their children. That is, if the
trait is fixed, motivation to control events related to that trait may be inhibited. Thus,
Dweck’s theory and the complexities is may address might be investigated after
additional initial studies regarding attributions and ADHD.

In summary, although Dweck’s research applies to perceptions of others, the
decision in the current study was to examine parent beliefs about ADI-ID at the attribution
level rather than at the level of Dweck’s implicit theory. While Dweck’s work has

implications for the study of parents’ goals and behaviors toward their children, the focus

l9

of the current study is on parental adjustment. Further research examining parental
behaviors may consider the implicit theories parents hold about their child’s ADHD
building upon potential results of the current study. Despite this, it is noted that Dweck
and colleagues (Levy & Dweck, 1998) found that individuals who ascribe to an entity
theory, tend to make more stable dispositional, stable trait inferences and to explain the
causes of behaviors in terms of an individual’s traits rather than to circumstances or goals
whereas incremental theorists tend to explain causes of a person’s behavior in terms more
situational factors. These findings support the co-occurrence of stable, uncontrollable,
global attributions for events if ADI-ID behavior is seen as uncontrollable and fixed.

Thus, the social cognitive literature lends theory as to what attributions may be
important in studying parental beliefs about their child’s ADHD behaviors. Notably, all
three primary theories reviewed suggest that stability and globality would be associated
with poorer maternal adjustment. The roles of locus and control are less clear. Learned
helplessness theory suggested that attributions that internal and uncontrollability
attributions would be associated with poor adiustment whereas Weiner’s research
indicated that internal and controllable attributions were associated with more negative
muse from help-givers. Thus, current study hypothesis pertaining to parents
attributions about child disruptive behaviors in relation to parent adjustment were
exploratory but of utmost interest.

Given this literature, it is useful to understand findings in relation to parents and
their children. Only a small literature has examined attributions in parents of children
with ADHD; thus, consideration is given to the growing literature which links parent

attributions to parental emotional and behavioral responses to their children generally.

20

Maternal Attributions
Maternal Attributions and Adjustment in Community Samples

In recent years, family and developmental psychologists have begun to study
parent attributions in relation to parental affective and behavioral responses to their
children (Bugental, Johnston, New, & Silvester, 1998; Dix & Grusec, 1985; Miller,
1995). In a review, Miller (1995) delineated both what “attribution” generally refers to
and also why attribution theory may be of interest to persons working with children and
families. First, although attributions have been operationalized in varying ways, Miller
purports that attributions refer to “the ways in which we explain and evaluate behavior”
(Miller, 1995, p. 1557). According to Miller and others (Johnston & Freeman, 1997;
Smith & O’Leary, 1995; Wright, Partridge, & Williams, 2000), parent attributions may
influence the ways parents affectively and behaviorally respond to and interact with their
children and may affect treatment decisions (Wright et al., 2000). In order to ease
readability, key empirical findings are summarized in Appendix A.

Studies have been conducted with parents of nonselected children with presumed
normal range behavioral adjustment, filrther elucidating the importance of the “learned
helplessness” dimensions in determining parental adjustment to their children’s behaviors
in general. Dix and Grusec (1985) found that the more parents of non-disordered
children viewed children’s behavior (in vignettes) to be internal and controllable (by the
child), the more upset the parents reported they would be. In a series of studies, Dix and
Grusec (1985) found that parents viewed self-control as internal and controllable

whereas lack of effort and lack of ability were viewed as internal and uncontrollable.

21

Misbehaviors that parents viewed as internal, controllable, and dispositional (stable)
were rated by parents as likely to be upsetting to them. Dix and Grusec concluded that
“self-control problems are, thus, upsetting and important to respond to” (1985, p. 224).
This work suggests that attributions that children’s behaviors are internal and
controllable are likely to be associated with increased parent stress if coupled with high
levels of child misbehavior. The existing research suggests that parent role adjustment
may be linked to parental attributions and that this relation might be mediated by actual
child misbehaviors. Taken together the studies with adults and parents of non—disordered
children suggest that inte;n_al. controlirble. a_ng§t_able Jattributioni for misbehaviors are
associated with increased parental stress.

An additional study, using a convenience sample of parents (through a mailed
survey sent randomly to unselected community participants), Geller and Johnston (1995)
found that on self-reported measures, attributions of more intemality and controllability
for child non-compliance was related to less investment in parenting and more intense
parenting responses. 1

In summary, attributions appear to be related to parents’ emotional responses to
their children. Further, according to Dix and Grusec’s (1985), attributions related to
“self-control” problems are of particular relevance to parental adjustment. This work
suggests that parent attributions about children’s misbehavior are likely to be
significantly related to parent role-stress. Because so little research exists on attributions
made by parents of children with ADHD, literature on attributions pertaining to other
externalizing behaviors (namely, aggressive behavior and conduct problems) pertaining

to controllability is first reviewed.

22

Attributions in Parents of Children With Behavior Disorders.

Parental attributions may differ for parents of children who exhibit substantial
problem behaviors compared to parents of non-disordered children with average range
problems. For example, mothers of children who display non-compliant, oppositional, or
aggressive behaviors (in some studies meeting diagnostic criteria for Conduct Disorder),
rated their children’s misbehaviors as more intentional compared to mothers of non-
disordered children (Baden & Howe, 1992; Bickett, Milich, & Brown, 1996; Dix &
Lochman, 1990; Strassberg, 1995). For example, Strassberg (1995) found that mothers
of oppositional, non-compliant boys made greater attributions of defiant intent compared
to mothers of non-disordered boys. Strassberg’s (1995) findings related to mothers of
disordered children contrast with Gretarsson and Gelfand’s (1988) findings pertaining to
mothers of non-disordered children. Mother’s of non-disordered children attributed child
misbehaviors to external, uncontrollable, and transient factors whereas child positive
behaviors were attributed to internal, controllable, and stable factors.

One particularly interesting study (Bickett, Milich, & Brown, 1996) found that
mothers of aggressive boys were more likely than mothers of non-aggressive boys to
infer negative motives and dispositions when explaining their sons’ behavior. These
mothers did ngt make more negative attributions when explaining a hypothetical child’s
behavior, they only did so when discussing their own son’s behavior. These mothers
exhibited a “hostile attributional bias,” a phenomenon well-established in aggressive
children (Crick & Dodge, 1994; Dodge, 1985). It represents a tendency of aggressive

youth to attribute hostile intent in ambiguous social encounters (Dodge, 1985). Bickett,

23

Milich and Brown’s (1996) study broke new ground in this line of research, finding that
not only did aggressive boys in their study exhibit a “hostile attributional bias,” but their
mothers did as well. Bickett, Milich, and Brown (1996) suggested that this bias may be
exhibited first by mothers and modeled to their sons. Because Bickett, Milich, and
Brown’s study was correlational, direction of effects is uncertain. Although parents of
aggressive children exhibited biased attributional styles, it is uncertain, how these
attributions relate to parental adjustment.

In a study comparing mothers of conduct-disordered and non-conduct disordered
boys, Baden and Howe (1992) found that mothers of boys with conduct-disorders were
more likely to attribute their children’s misbehavior to stable and global causes and more
likely to see the behavior as beyond the parent ’s control. Mothers of disordered and non-
disordered children did not differ in the extent to which misbehavior was seen to be
controllable or uncontrollable by the child. Because the internal consistency of the locus
subscale (internal vs. external) showed that scale to be unreliable, Baden and Howe did
not analyze that attributional domain. This study suggested that mothers of conduct-
disordered children believe the causes of their children’s behavior to be stable, global,
and uncontrollable. It may be these attributions serve a self-protective function,
preserving the parents self-esteem because although they see the child’s behaviors as
stable and global they are seen as not within parental control. Alternatively, they may be
depressongenic attributions; parents may be giving up control due to failed attempts at
control.

Baden and Howe (1992) also examined parents’ expectations about the

effectiveness of their parenting behaviors, finding that parents of disordered children felt

24

their parenting was less effective. The causal pathway of these relations is unclear. It is
possible that attributions about parenting and their child’s behavior are made after
persistent negative encounters with the child’s disruptive behavior or that parental
attributions and stress contribute to children’s aggressive behavior akin to a “self-
fulfilling prophecy.” A self-fulfilling prophecy would exist if parent’s beliefs led to
particular parent behaviors that in turn resulted in the child behaving in such a way that
fulfilled those parental beliefs.
A Self-Fulfilling Prophecy

Although not a study of parent attributions, a recent study of unacquainted
children suggests that at least amongst peers, children may have expectations about the
behavior of someone diagnosed with ADHD. In a study of 68 pairs of unacquainted boys
in Grades 3-6, Harris and colleagues (Harris, Milich, Corbitt & Hoover, 1992) examined
interactions following a belief manipulation. In each dyad, a non-disordered child was
told that his partner had a behavioral problem or was given no information that might
lead to a preformed attributional expectancy. Sometimes the target child actually had
ADHD and other times the child did not. Interactions were more disruptive when the
target child had ADHD 9; when the target child had been identified as having a behavior
disorder (even when the child did not hip/e such a disorder). This study demonstrated
that labeling a child as ADI-ID adversely affected peer interactions even when the labeled
child did no_t truly have ADHD. These findings suggest that negative attributions or
expectations about ADHD may result in worse behavior from the target child. The
direction of effects is unknown. It cannot be concluded whether the target children

adjusted their behavior to meet expectations or whether the other children treated the

25

labeled children differently precipitating different behavior. If children with ADHD are
treated differently due to labeling effects, they may adjust their behavior accordingly.
Their beliefs about themselves may change as well. Some researchers (Bickett, Milich,
& Brown, 1996; Strassberg, 1995) suggest that mothers of aggressive children may make
attributions that result in an increase in their child’s misbehaviors in a “self-fulfilling
prophecy” fashion.

This line of research points to the importance of examining parental beliefs and
attributions about children’s ADHD diagnosis and children’s misbehavior. Because we
know that children with ADI-II) are at risk for developing aggression (Anderson et al.,
1987), the examination of attributions made by parents of children with ADHD in
relation to their children’s misbehaviors is a logical extension of the work with parents of
conduct-disordered children. Currently, a few studies have begun to investigate

attributions made by parents of children with ADHD.

Attributions made by Parents of Children with ADHD
The study of attributions made by parents of children with ADHD is a relatively

new area of research. Most studies of attributions in relation to ADHD have focused on
attributions made by the child (Rosen, O’Leary, & Conway, 1985). Indeed, relatively
few studies have examined parent attributions about children in families with an ADHD
child, yet parental attributions are thought to affect parents’ adjustment, behaviors toward
their child, and treatment choices pertaining to their child’s ADHD (Edwards, Schulz, &
Long, 1995; Hoza et al., 2000; Wright, Partridge, & Williams, 2000) as well as parenting

behaviors in general (Hoza etal., 2000; Slep & O’Leary, 1998), general parental

26

adjustment (Podolski & Nigg, 2001) and outcomes for children with ADHD (Hinsbaw et
al., 1997; Hoza et al., 2000).

Findings pertaining to parental attributions about their child’s ADI-ID vary.
Johnston and Freeman (1997) and Johnston, Reynolds, Freeman, and Geller (1998) found
that parents of children with ADHD rate hyperactive and oppositional behaviors a_s less
controllable by the child, and more stable compared to ratings made by parents of non-
disordercd children. Sobol, Ashboume, Barn, and Cunningham (1989) found that
mothers of children with ADHD rated the child’s behavior as less stable compared to
mothers of non—disordered children. Both groups of researchers found that parents of
children with ADHD generally believe that their child’s misbehavior is not within the
child’s control. In fact, Sobol et a1. (1989) found that parents of children with ADHD
believed that their own (parental) role in “causing” (as measured by causal attributions)
the child’s misbehavior was unstable and uncontrollable. That is, whereas parents of
non-disordered children apparently feel responsible for their child’s misbehavior
(Johnston & Freeman, 1997), parents of children with ADHD believed that due to
ADI-ID, their child’s behavior was out of both the child’s and the parent’s control
(Johnston & Freeman, 1997; Sobol, et al., 1989).

As suggested by the study with unacquainted peers (Harris et al., 1992),
attributions made parents of children with ADHD may exemplify the power of the
diagnostic label to influence beliefs. Alternatively, parental attributions may be driven by
experiences with their child’s misbehavior over time. Notably, attributions that children
with ADHD cannot control their misbehaviors are consonant with prevailing views about

the biogenetic nature of ADHD.

27

Findings pertaining to parent attributions about locus of control (internal vs.
external) are unclear and differ based on child’s medication status. When comparing
parents of children with ADHD to parents of non-disordered children (without
medication status controlled or examined), Johnston and Freeman (1997) found that
parents of children with ADI-ID believed inattentive-overactive and oppositional defiant
behaviors were more internally caused, less controllable, and more stable. When
comparing maternal attributions for child behavior when the child was medicated versus
not medicated, Johnston, Fine, Weiss, Weiss, Weiss, and Freeman (2000) found that

mothers of children with ADHD rated their child’s negative behaviors as more externally

 

caused when the child was medicated compared to when the child was unmedicated.
These findings once again suggest that attributions made by parents of children with
ADHD coincide with what might be expected given the prevailing biogenetic model of
ADHD (Tannock, 1998).

Notably, the attributions made by parents of children with ADHD may exemplify
a self-esteem preserving attribution by parents of difficult children. Or they may simply
be an accurate reflection of reality. In any event, these findings indicate the need to
develop further consideration of how these parents view child problems. If self-
protective, one might expect them to be related to lower parental role-stress. However, if
an accurate reflection of reality, they may be related to increased role-stress.
Consideration of attributions made by parents of children with ADHD in relation to their

role-adjustment is a logical next step.

28

Attributions and Adjustment When a Child has ADHD

Two recent studies provide links to parent-role adjustment. In a recent study of
100 mothers and 57 fathers of children with ADHD, Hoza and colleagues (2000) found
that both mothers’ and fathers’ attributions of non-compliance to child’s lack of effort
wararelatg to low parenting efficacy. Although they did not compare parents of
children with ADHD to parents of non-disordered children, this study provided evidence
of a link between parental attributions and self-reported emotional response.

Hoza and colleagues (2000) also investigated parental attributions in relation to
parenting behaviors and child treatment outcome as part of a study of ADHD
interventions. As noted earlier, research with non-ADHD samples suggests that parent
attributions are likely to be related to parental adjustment, parenting behaviors, and child
behaviors. In the study with ADHD sample, Hoza and colleagues (2000) found that
parental attributions about their children were significantly and strongly (r=.83, p<.01 for
mothers; r=.29, p<.05 for fathers) related to discipline. Specifically, for both mothers and
fathers, attributiona of child misbehavior to low child effort were related to dysfunctional
discipline strategies by parents. Although the relation between parent attributions and
child treatment outcomes was less clear, Hoza found that parental adjustment (maternal
low self-esteem, paternal low parenting efficacy) predicted worse child treatment
outcomes in a subset of the Multimodal Treatment Study of ADI-II). Their study
suggests that both (a) parent attributions about their child and (b) parental role—adjustment
may be related to child outcome.

Although not examining attributions about the child’s misbehavior per se, a recent

study by the current author found that parents’ self-reported cognitive reframing of

29

family problems related to their child with ADHD was related to lower parental stress
levels and mediated the relation between child misbehavior and parental role-adjustment
(Podolski & Nigg, 2001). This finding suggested that attributions play an important role
in parental adjustment to children’s ADI-ID. Specifically, when parental attributions are
associated with the possibility of efficacy behavioral responses, parents are likely to
experience lower levels of stress. However, given the nature of ADHD, it is possible that
parents may believe that their child’s behaviors are not controllable. Following the above
reasoning, this might be related to an increase in stress. The current study examined
specific attributions about the child’s misbehaviors in relation to parental adjustment in
order to elucidate additional cognitive attributions that might be related to parent role-

stress.

Summary and Critique

The few existing studies of attributions made by parents of children with ADHD
indicate that parents believe that when a child has ADI-ID his/her misbehaviors are
lnLemallv caused, uncontrollable bv the child, and stable (Johnston & Freeman, 1997 ;
Johnston et a1. 2000). How these attributions are related to parental adjustment remains
unclear. While studies of parents with non-disordered children have indicated that
internal, controllable, and dispositional (stable) attributions for children’s misbehaviors
are most upsetting to parents (Dix & Grusec, 1985), Hoza et al.’s (2000) study with
parents of ADHD children suggests that internal and uncontrollable attributions (as
measured by “low effort attributions” are related to low parenting efficacy. It is possible
that the relation between attributions and adjustment is moderated by child diagnosis (or

severity of child behaviors). Notably, Hoza et al. (2000) only assessed two domains of

30

attributions (effort and locus of control). Although theorists (Dix & Grusec, 1985) have
speculated that low effort attributions are equivalent to internal and uncontrollable
attributions, this remains to be tested in ADHD samples. A major weakness in the
existing literature is the failure to examine parent role-adjustment in relation to multiple
domains of attributions.

The current study attempted to replicate and further Johnston’s study regarding
the pattern of attributions made by parents of children with ADHD and to explain Hoza et
al.’s (2000) findings regarding possible relations between parent attributions and
adjustment. Irnportantly, the proposed study tested the relation between (a) parent
attributions and (b) parent role-stress, with severity of child disruptive behaviors
controlled. This had not been done in prior studies and yet was considered important in
order to test whether parent attributions about child behaviors are independently related
to parental adjustment above and beyond severity of child behavioral problems. It was
hypothesized that parental attributions about their child’s misbehaviors would be related
to role-stress above and beyond that accounted for by their child’s misbehaviors (as
measured by parent ratings; see method). Because the diagnostic label also might have a
significant role in predicting parent adjustment, the study also aimed to test whether child
diagnosis moderated a potential relation between child behavior and parental adjustment.
Thus the study aimed to explicate not only whether parental attributions were related to
parental adjustment to their children’s disorder but also possible pathways of this

relation.

31

CHAPTER 2
LITERATURE REVIEW FOR STUDY 2: CHILD ATTRIBUTIONS

A second exploratory portion of the study was to examine child attributions.
Through modeling and indirect instruction, children may learn schemata about their own
behaviors through parents’ actions and statements. Thus, children may learn attributions
about their own behaviors from hearing their parents verbalize attributions. A primary
complication in understanding attributions in relation to ADI-II) is that the literature
suggests that attributions differ depending on diagnosis. Specifically, as reviewed in the
prior section, the literature on parental attributions suggests that attributions in parents of
children with ADHD may operate differentially compared to the way they function in
parents of children without a disorder. The literature on child attributions suggests
similarly that the relation between attributions and adjustment may differ for children
based on their ADHD status, with the relation differing from what has been found within
the general attribution literature.

Children’s attributions may be linked to their adjustment in several ways. For
example, as will be reviewed in more detail, given the relation between “helpless
attributional styles” and helplessness in adults it was expected that a similar pattern of
“helpless” attributions and behaviors would be associated with worse adjustment in
children. Early research of attributions in children found some surprising results.
Additionally, when these attributions were studied in children with ADHD, the findings
were mixed. As outlined below, when examining attributions made by children with
ADHD, some authors outlined a pattern of “helpless” attributions and academic

behaviors whereas others argued that a similar style may serve a self-protective function

32

in children with ADHD (Carlson, Mann, & Alexander, 2000; Johnston, et al. 2000;
Milich, 1994). The early studies of “helplessness” in children will be reviewed followed

by a review of the initial studies of children with ADHD.

“Helplessness”and “Mastery” Attributions.

Some of the initial research on children’s attributions was conducted by Dweck
(Dweck & Reppucci, 1973) and examined attributions of responsibility pertaining to
academic tasks. These early studies with non-disordered children suggested that external
attributions of responsibility were related to quitting behaviors. These findings
contrasted with the learned helplessness literature that internal locus of control was
related to helplessness when adults faced failures.

Using attributions about responsibility, Dweck (Dweck, 1975; Dweck, Goetz, &
Strauss, 1980; Dweck & Reppucci, 1973) reported that children who were gave up in the
face of failure, “took less personal responsibility for both successes and failures.” This
“EM” behavior corresponded with “M” rather than “internal” responsibility
attributions. The amount and sequencing of successes and failures was experimentally
manipulated so that the children received the same experience (in terms of number and
order of successes and failures on the tasks). Despite the experimental manipulation,
differences in behavioral responses and attributional styles were found. As noted above,
“helpless” behaviors (lapkof effort) were associated with e_xtemal attributions. In
contrast, children who made internal responsibility attributions were more likely to

respond with effort and were termed “mastery-oriented.”

33

Notably, Dweck’s findings differed fiorn the general “learned helplessness”
model in adults (Peterson, Maeier, & Seligrnan, 1993). Whereas Dweck found “external”
attributions to be associated witlgraasivitv. Petersyglhad found “internal” attributions for
naggive event; to often be associated with helplesm. (Internal attributions for
positive events were associated with optimism and good adjustment).

A few noteworthy differences between the studies may-have accounted for the
different results. First, the events in Peterson’s early studies were clearly
“uncontrollable” whereas in Dweck’s study controllability was not directly assessed. In
Peterson’s later studies, in using events that could be deemed either controllable or
uncontrollable by the participant; Peterson still found that internal attributions for
negative events were associated with depression (Peterson & Barrett, 1987).
Additionally, adult learned helplessness studies examined attributions for positive and
negative events separately, finding that “intern ” attributions for negative events were
often associated with depression and that internal events for positive events were
consistently related to good adjustment. The differences between Dweck and Peterson’s
findings indicate that other factors must be considered in addition to the internal versus
external dimension of attributions and that the valence - positive versus negative - of the

events considered is important.

Performance Attributions by Children With ADI-ID.
Because children with ADHD experience numerous academic fi'ustrations
(Barkley, 1998), it is plausible that their academic failures may be related not only to

poor attention but also to a “learned helpless, pessimistic” belief system and response

34

style. Although beliefs and attributions held by children with ADHD have been

examined, the findings offer mixed results.

Helpless Style
In a study of the effects of reward and response cost (as a punishment), Carlson,
Mann, and Alexander (2000) found that children with ADHD endorsed a “less adaptive
atttributional style” compared to non-disordered children. Carlson and colleagues (2000)
found that children with ADHD (n=40; 27 boys, 13 girls) were more likely than matched
non-disordered children (n=40) to attribute positive events to luck (external,

uncontrollable) and poor performance to lack of effort (internal, controllable). The

 

children with ADHD made less internal stable, and global attributions for positive events
and more internal, stable, and global attributions for negative events relative to non-
disordered children. Such explanations are i_nconsistent with Peterson’s optimistic
explanatory style1 but consistent with Peterson’s pessimistic explanatory stylez.
Unsurprisingly, Carlson and colleagues concluded that children with ADHD displayed a
“less adaptive attributional style.”

Carlson et al.’s (2000) findings also are consistent with a learned helplessness
response. Notably, Carlson also found that children with ADHD were less likely to make
internal and more likely to make external attributions for positive events, a pattern which
is opposite of an “optimistic explanatory style.” Carlson et al’s (2000) findings also
indicated that attribution/behavior pattems of the children with ADHD were similar to

those exhibited by children with learning disabilities. A study on attributions in children

 

' Optimistic’explanatory style: Stable, internal, specific attributions for positive events. May also
involve: external, unstable, and specific attributions for bad events.

35

with learning disabilities found that internal attributions for poor performance were
associated with decreased effort (Licht et al., 1985) in contrast to the “mastery” behaviors
associated with internal attributions in non-disordered children (Dweck & Leggett,
1988). Also, when asked about how well they thought they performed, children with
ADI-II) also reported poorer performance compared to controls (Carlson et al., 2000). In
terms of actual performance, children with ADHD attempted fewer problems, got fewer
problems correct, and received a lower percent correct ratio compared to control children.
Hence, Carlson et a1. (2000) found internal attributions for negative events and arisrpal
attribution_s for pogive events to be associated with poor performance in children with
ADHD on an academic task.

Similarly, Hoza, Pelham, Waschbusch, Kipp, and Owens (2001) found that
ADHD boys endorsed attributions that were more external for successes and less internal
(more external than controls even though still internal attributions) for failures. Hoza and
colleagues (2001) found the ADHD boys performed worse and were rated as more
uncooperative and less effortful. They solved fewer puzzles and quit working more
often. Hence, Hoza et a1. concluded that external attributions were associated with less
effort. Notably, Hoza et al.’s finding that internal attributions for failure were endorsed
less strongly by ADHD boys compared to controls is consistent with a possibly protective
attributional style. [Note as shown in Appendix B, according to Peterson (Peterson,
Maier, & Seligrnan, 1993), internal attributions for failure are associated with sadness
and depressed mood; see Appendix A]. However, Hoza et al.’s finding that ADHD boys
made more external attributions for successes is inconsistent with an optimistic

explanatory style and instead likely consistent with helplessness so they may show both a

 

2 Pessimistic explanatory style: stable, internal, global, uncontrollability attributions for negative

36

protective and helpless bias depending on context. Hoza and colleagues also included a
measure of the child’s disruptive behavior. Research assistants rated the boys levels of
effort and cooperation during the tasks. ADHD boys were rated as being less effortful

and cooperative, but the ratings were not related to the boys’ attributions.

Protective Style

In contrast to the idea that children with ADHD endorse “helpless” attributions,
Milich (1994) has argued that external attributions for poor performance may be adaptive
in children with ADHD. Whereas Dweck and Leggett (1988) found external attributions
for poor performance to be associated with “helpless” behaviors in non-disordered
children, Milich (1994) did not find external attributions for poor performance to be
related to “helpless behaviors” in children with ADHD. Notably, according to Peterson’s
learned helplessness theory, internal attributions for poor performance are related to
depression (Peterson & Barrett, 1987; Sweeney, Anderson, & Bailey, 1986). Milich
(1994) reviewed multiple studies conducted by himself and his colleagues and concluded
that external attributions were not related to poor adjustment in children with ADHD.
Milich (1994) argued that although many children with ADHD experience greater
frustration and may give up, these “helpless” behaviors may not be associated with
“helpless” attributions. Rather Milich (1994) argued that what are commonly seen as
“helpless” attributions may be related to more beneficial coping in children with ADHD.
This program of research is next summarized.

Milich and Okazaki (1991) put Dweck’s hypotheses (that external attributions

were related to helplessness and internal to mastery) directly to the test in children with

37

ADI-ID. They found that boys with ADHD solved fewer word puzzles than non-
disordered boys, gave up more often, and reported more frustration. Thus, some support
was found for “helpless” behaviors in boys with ADHD. However the “helpless”
attributional style as defined by Dweck did not appear to operate in boys with ADHD as
would have been predicted. Using a median split on the Intellectual Achievement
Responsibility Scale (Crandall, Katkovsky, & Crandall, 1965), Milich and Okazaki
(1991) found that boys who made external attributions for failure also reported trying

harder when compared to the boys who made efi’ort (internal) attributions.

 

Milich and Okazaki’s (1991) findings appear to contrast with Dweck and
Leggett’s (1988) finding that boys with high effort attributions worked more persistently
and had a “mastery-oriented” style whereas boys with lo_w effort (and high external)
attributions worked less persistently and displayed a “helpless” style. Notably, Milich’s
findings are not laconsistent with Peterson’s (Peterson, Maier, & Seligrnan, 1993). The
latter found that internal attributions for positive events were related to good adjustment
but that internal attributions for negative events are related to poor adjustment in non-

disordered adults. In fact, Milich and Okazaki’s (1991) findings of external attributions

 

for 113ng events are more consistent with Peterson’s “leamed helplessness” than are
Dweck’s findings that external attributions are related to helplessness. It is also
noteworthy that when using a median split Milich found that some boys with ADHD
made internal attributions for failure whereas some boys with ADI-ID made external
attributions for failure. Unlike Carlson et a1. (2000), Milich and Okazaki (1991) did not

compare children with and without ADI-ID. Milich and Okazaki (1991) also did not look

38

at different attributions _vyi_pthi_p the ADHD group, or compare attributions made by boys
and girls.

Milich (1994) proposed that, for children with ADHD, external attributions or
tailures may be more adaptive than ipternal attributions in contrast to Dweck’s
conclusions based on finding in non-disordered children that internal attributions (for
both failure and successes) were related to more effort. Indeed, in the Milich and
Okazaki (1991) study, Milich also found that boys who made effort attributions for
failure (internal attributions) were more likely to perform like “helpless” children,

whereas ADHD boys who made external attributions tor tailure exhibited an adaptive

 

behavioral response. Milich (1994) concluded that attributions that may be deleterious
for non-disordered children may be adaptive for children with ADI-II). Recall that the
internal attributions for negative events were part of the “pessimistic explanatory style”
elucidated by Peterson (Peterson, Maier, & Seligrnan, 1993), who examined attributions
for ancontrollable events. Once again, it seems that whether the individual rates the
event as uncontrollable and external versus uncontrollable and internal may be important.
Milich and other researchers have also compared attributions and persistence
behaviors of children with ADHD when taking medication for ADHD to their attributions
and behaviors in a placebo condition (Carlson, Pelham, Milich, & Hoza, 1993; Milich,
Carlson, Pelham, & Licht, 1991; Milich, Licht, Murphy, & Pelham, 1989; Pelham et a1.
1992). In these studies, Milich and colleagues consistently did not find evidence that
medication was associated with handicapping attributions. For example, Milich, Licht,
Murphy, and Pelharn (1989) studied ADHD boys’ attributions for performance when on

medication and when on placebo, finding that r_eg_ardless of medication stiLua, the ADHD

39

boys attributed success to own their own effort. and they did not use medication as an
explanation for successes. This study further supported Milich’s notion that ADHD boys
did not make deleterious attributions. Rather boys seemed to make internal and
controllable attributions for positive events and external (and possibly, although not
tested by Milich, uncontrollable) attributions for negative events. Thus, Milich (1994)
concluded that ADI-ID boys may hold a self-benefiting/protective bias.

To summarize, the bulk of the studies on children with ADHD found external

 

attributions for failure an attributional style consistent with better adjustment. These
studies are in contrast with Dweck’s definition of helplessness in non-disordered children
but consistent with Peterson’s theory about depression. External attributions for

successes however, were not consistent with positive adjustment. Finally, studies of

 

children with ADHD differ in their findings in regard to locus attributions. Milich and
colleague’s work suggests that boys with ADHD made internal attributions for success
but Hoza and colleagues found opposite results.

Notably, the literature reviewed above pertains to child beliefs in relation to
academic type tasks. These studies did not examine child attributions in relation to co-
morbid problems, such as aggression, so common in children with ADHD. Given that
attributions are likely to affect child personal and social interactions, these non-academic
outcomes deserve attention and may elucidate results pertaining to parent as well as child

attributional ratings of child disruptive behaviors.

40

Attributions About Disruptive Behaviors

A body of literature has examined attributions about misbehaviors made by
children who act out aggressively (Dodge & Coie, 1987 ; Hudley & Graham, 1993) and
by children with ADHD (Johnston et a1. 2000). Children who are aggressive tend to
view others aggressively (Dodge & Coie, 1987 ; Graham, Hudley, & Williams, 1992;
Hudley & Graham, 1993). Also, studies find “halo effects” between ratings of defiance
and hyperactivity. For instance, Schachar, Sandberg, and Rutter (1986) found that when
a child behaved defiantly, he was also likely to be rated as hyperactive or inattentive
regardless of his observed level of activity. Aggression is the most common co-occurring
problem for children with ADHD and is associated with poor short and long-term
outcome. Despite these findings, few studies have examined beliefs or attributions about
disruptive behaviors made by children with ADHD.

Because aggressive and oppositional/defiant behaviors are as important (or
possibly more important) as ADI-ID symptoms in predicting the long-term outcome of
children with ADHD ( Hechtrnan & Weiss, 1983), some studies have begun to measures
children’s attributions about these behaviors. Two known studies have measured
children’s attributions about ADHD and aggressive/defiant behaviors.

First, in a study of 74 non-disordered children, Johnston and Leung (2001)
examined attributions about a video-taped child’s ADHD behaviors. Children were told
that the child was receiving either: a) no treatment, b) medication only, c) medication
plus behavioral treatment. The children (all boys) saw ADHD behaviors as m
controllable by the child when treatment involved a behavioral component.

Noncompliance was seen as more intentional and marginally more controllable if the

41

child was receiving treatment but controllable when treatment involved behavior
management. Child self-attributions were not assessed.

Secondly, using a sample of 86 children with ADHD, Johnston and colleagues
(Johnston et al. 2000) examined the attributions about their own behavior in relation to
their medication treatment. These children rated their compliance and noncompliance as
more controllable when they were on medication. Also, Johnston et a1. (2000) compared
children’s attributions on a forced-choice measure and found that children ascribed
compliance to ability, effort, and the task for their behaviors off-medication but ascribed
compliance when on medication to “pill-taking.” These findings appear to mirror the
early anecdotal reports (Rosen, O’Leary, & Conway, 1985; Whalen & Henker, 1991).
The effects of ascribing compliance to medication are unknown but some theorists
purport that such attributions might have negative effects on children’s self-esteem.

Although not a study of attributions per se, the study of self-esteem is one way in
which beliefs about the self has been assessed in children with ADHD. A prospective
study of adolescent hyperactive children (Slomkowski, Klein, & Mannuzza, 1995) found
that hyperactive boys reported lower self-esteem as adolescents, had lower educational
and occupational ranks as adults, and had lower overall adjustment compared to non-
disordered control subjects. Although not examining beliefs about ADHD, these
hyperactive individuals did have negative self-perceptions as measured by a self-esteem
measure. As is commonly the case in ADHD studies, information on girls was not
available.

Despite this finding, other studies indicate that children with ADHD do not have

lowered self-perceptions (Hoza, Pelham, Milich, Pillow, & McBride, 1993). For

42

example, in a study of 27 boys with ADHD (with and without CD combined, no
differences found when examined separately) and 25 boys without disorders (ages 8.5 to
13), Hoza and colleagues (1993) did not find differences in self-perceptions and
attributions when comparing boys with and without ADHD. Notably, seventy percent of
the boys with ADHD had internalizing symptoms that were clinically significant
compared to only 20% of the non-disordered boys. Due to the presence of internalizing
symptoms in the ADHD sample, the authors compared self-perceptions and attributions
both without and with controlling for internalizing symptoms. Hoza and colleagues
found that boys with ADHD were more likely to attribute msitive social outcomes to
themselves and less likely to attribute negaliveiltcomesy to internal causes; these
findings held even when controlling for internalizing symptoms. Such attributions
would be consistent with a “self-protective” bias. However, in contrast to a “self-
protective” attributional style, boys with ADHD were more likely to make stable and
global attributions for negative events: when controlling for internalizing symptoms, the
boys with ADI-II) were still found to make stable but not global attributions for negative
events. This study further underscored the importance of examining factors in addition
to the intemal/extemal locus. Due to low power, results must be interpreted with

caution.

Critique of Current Findings with ADHD Children
Results pertaining to attributions made by children with ADHD are mixed. A
series of studies by Milich (Hoza, Pelham, Milich, Pillow, & McBride, 1993; Milich,

Carlson, Pelham, & Licht, 1991) suggests that external attributions in ADHD may be

43

helpful (self-protective) in contrast to what is found with non—disordered children.
However, more recent research suggests that children with ADHD make attributions
which are consistent with learned helplessness and with “helpless” behaviors ( Carlson et
al., 2000; Hoza et al., 2001); for example, Carlson et al. (2000) found that boys with
ADHD made internal attributions for poor performance. The lack of consensus within
the literature points to the need for additional studies.

Most studies examining attributions in children with ADHD, including those by
Milich (1994), Carlson et al. (2000), and Hoza et al. (2001), did not measure attributions
pertaining to controllability. Most studies, including those by Milich (1994) and by Hoza
et al. (2001) focused on the locus (internal vs. external) dimension, failing to measure
stability, globality, or controllability. Recent work by Peterson and Bossio (1991) has
found that internal locus for bad events is not consistently related to negative adjustment.
For example, Peterson and Bossio (1991) found that stability and globality but not
intemality was related to poor health outcomes. Given these recent findings, the failure
of attribution studies to include measures of stability and globality are serious. In fact,
the mixed findings pertaining to attributions made by children with ADHD may be
explained through these unmeasured dimensions. Additionally, most studies of children
did not assess the extent to which events or behaviors was seen as “lmeontrollable” yet
this dimension appears crucial to the definition of “learned helplessness” put forth
originally and likely plays an important role in all findings.

Additionally, the current literature indicates that whether internal or external
attributions are related to positive outcomes for children with ADHD may depend on

specific contexts and the specific behaviors ( academic, general ADHD behaviors,

reading tasks, use of cognitive self-regulation and focus) being studied. Examination of
actual disruptive behavior may be of greatest importance to understand long-term
outcome; this outcome is also the least well studied. Presently, very few studies have
examined attributions made by children with ADHD in relation to common co-morbid
problems, such as aggression. Given that attributions are likely to affect child personal
and social interactions, these non-academic outcomes deserve attention.

An additional common weakness in the ADHD literature and in attribution studies
is the exclusion of the study of girls. To date, very few studies have examined the
attributions of girls with ADHD. When studies included girls, gender may not have been
analyzed due to limited sample size (Carlson, Mann, & Alexander, 2000). When
analyzes were conducted separately for boys and girls (in non-ADHD studies), sex
differences were found (Dweck, Goetz, & Strauss, 1980).

In conclusion, perhaps the most striking weakness in the literature is that despite
contradictory findings regarding internal /extemal locus, most studies failed to assess
other attribution dimensions, such as controllability, globality, and stability. Yet these
factors have been found to discriminate depressed and non-depressed individuals
(Peterson, Maier, & Seligrnan, 1993). Another central short-coming of existing literature
is the relative paucity of studies examining attributions in relation to non-academic tasks
in children with ADHD. This is a relatively new yet important line of research that
merits further exploration. Notably, child aggression is important to study and arguably

may be tied to children’s attributions about their own disruptive behaviors.

45

Medication Effects

Reports of “deleterious” effects of beliefs in children with ADHD began with case
studies about their beliefs concerning medication (Rosen, O’Leary, & Conway, 1985;
Whalen & Henker, 1976). Specifically, the children with ADHD reported believing that
their pill helped them behave well and feared that they could not do well without their pill
(Rosen, O’Learly, & Conway, 1985). In these case studies, researchers and clinicians
noted that children “worried about what would happen if they could no longer take pills”
(Whalen & Henker, 1991, p. 237) and in some cases asked for pills to help behave
(Whalen & Henker, 1976). Before experimental studies were conducted, therefore
theorists had posited as to ways in which attributions about medication and ADHD might
affect children.

In an early paper discussing possible socio-ecological effects of psychostimulants
for children, Whalen and Henker (1976) outlined some possible attributional messages of
medication. As part of this discussion, it is important to acknowledge the effectives of
psychostimulants in ameliorating behavioral problems. Psychostimulants are effective in
improving child restlessness, on-task behavior, compliance, and current classroom
performance (Barkley, 1998; DuPaul & Barkley, 1990). These positive effects of
medication are large and well documented (Abikoff, 1991; Greenhill et al., 2001;
Richters et al., 1995; Vitiello et al., 2001). Although these positive short-term behavioral
effects are indisputable, understudied is the socio-ecological message implicit in such
treatments that may affect children’s psychological development in the long-term.

According to Whalen and Henker (1976, p.1122), “medication prescribed for the

46

regulation of behavior . . .communicates to the child and to those around the child that a
definable problem exists, most likely a physiological one, and that it can be alleviated
through chemical means.” Whalen and Henker therefore hypothesized that stimulant
treatment facilitates external attributions about problem solutions, such that children are
not responsible for their failures or their successes. Such attribution; could haya
protective effects in terma of the child not blaming him/herself for failure, but could have
@terioas effecta if the children’s failure to engage in behavioral control strategies
within their capability and beneficial to their development resulted in 1n_app' ropriate

abdication of responsibility.

Conclusions from Literatures: Attributions and ADHD

Overall, the literature on both parent and child attributions indicate that the
attributions endorsed by mothers of children with ADHD (and by children with ADHD)
' may both differ significantly both in the attributions that are endorsed and in the function
of those attributions. The literature suggests that unlike mothers of children without a
disorder, mothers of children with ADHD attribute misbehaviors to more stable, global,
and uncontrollable causes whereas parents of children without a behavioral disorder tend
to attribute child behaviors to uncontrollable and transient factors. The locus (internal
versus external) dimension was less clear; however, parents of children with ADHD
appear to make more internal attributions. The relation between maternal attributions
about child misbehaviors and maternal adj ustment has not yet been explored in a sample
of parents of ADHD children using the domains primary attribution domains outlined

above. Notably, the pattern of attributions endorsed by mothers of children with ADHD

47

(Johnston & Freeman, 1997) differs from the pattern that has been associated with
parental stress in other samples of parents (Dix & Grusec, 1985). Thus, for both parents
and children with ADHD, the pattern of attributions “normally" (in non-disordered
samples) associated with poorer adjustment may be associated with better adjustment in
an ADHD sample. That is, the pattern of attributions associated with poor adjustment
may differ when considering ADHD. The controllability domain appears particularly
different from prior studies for both parents and children. Notably, absent from the few
existing studies is any examination of effects by subtypes of ADHD, both in terms of

patterns of attributions as well as in terms of the attribution-adjustment relation.

48

RATIONALE AND PLAN OF STUDY

As outlined in the review of the literature, attributions may be an important factor
in terms of maternal adjustment to their children’s ADHD. Notably, multiple domains of
literature within psychology point to the significance of attributions in adjustment.
Studies within social cognitive psychology have documented that attributions are linked
to a person’s evaluation of and responses to others (W einer, 1980). Studies within the
child development literature indicate that mothers’ attributions about their children’s
behavior are related to their affective and behavioral responses to that behavior (Dix &
Grusec, 1985). Given the documented high levels of stress experienced by mothers of
children with ADHD (Johnston & Mash, 2001; Podolski & Nigg, 2001) and the potential
implications for both mothers’ and children’s welfare, understanding factors which might
ameliorate or moderate that stress and it’s consequential results is imwrtant for
researchers and clinicians alike. In order to expand the existing literature and lend
knowledge which might be used for designing interventions, the current study aimed to
extend knowledge about maternal attributions in a sample of mothers whose children
have ADHD but also to examine these attributions in relation to mothers’ adjustment and
to child ADHD subtypes.

The proposed study had three overarching objectives. The first objective was to
extend our knowledge of maternal attributions about child disruptive behaviors
(hyperactivity and aggression). Attributions made by mothers of children with ADHD
were compared to attributions made by mothers of children without ADHD to see if
group differences existed (corresponds to hypothesis 1). In replication of Johnston and

Freeman (1997) it was expected that mothers of children with diagnostic levels of ADHD

49

would make more internal stable and uncontrollable attributions about children’s

 

disruptive behaviors compared to attributions made by mothers of non-disordered
children.

In a “logical next step,” the second objective was to examine specific maternal
attributions about their children’s misbehaviors in relation to mothers’ own role-specific
adjustment (satisfaction and stress; corresponds to hypothesis 2). Two competing
lambases wepe tested. Based on the literature, it was expected that the relation between
maternal attributions and role stress would differ based on child diagnosis (moderation by
child diagnosis). It was expected that mothers of children with ADHD would experience
greater levels of stress than control parents and that the relation between attributions and
adj ustment would be different for mothers of children with versus without ADHD.

Based on the general literature, it was hypothesized that internal stable and

 

controllable attributionjs would be related to gleater role stress among control parents.

 

Thus, with child behavior controlled, this attributional pattern was expected to relate to
maternal role stress (in replication of Dix & Grusec’s research with families whose child
did not exhibit a behavioral disorder (1985)). Notably, a different pattern of results was
expected in mothers of children with ADHD. Mothers of children with ADHD were
expected to attribute misbehaviors to uncontrollable causes whereas parents of children
without ADI-ID were expected to attribute misbehaviors to controllable causes. For
mothers of children with ADHD, role stress was expected to correlate with attributions of
Econtrollable behavior. Thus, the attribution pattern associated with role stress was

expected to differ for parents based on child diagnosis. Altematively was the possibility

50

that maternal attributions would be related to adjustment independent of child behavior.
Mediation by child behavior was also tested.

The _third obi ective was exploratory. This was to examine child attributions about
disruptive child behaviors, examining multiple dimensions of attributions about
misbehaviors (rather than academic performance as in prior studies). Using a newly
adapted measure of attributions that provides multiple dimensions, the aim was to
provide a more differentiated description of attributions and to examine non-academic
behaviors. Because the measure was newly adapted and a range of ages of children were
sampled, the primary expected contribution was the creation and piloting of the new
measure with the hopes of examining possible different attributions made by children
with and without ADHD. It was hoped that analyses would provide direction to future
studies aimed at clarifying contradictory findings of uni-dimensional studies and thus
extend the literature by examining attributions by children to real world disruptive child

behaviors.

51

CHAPTER 3
METHOD

Participants

Participants were: 126 children, 125 mothers, and 114 teachers. For diagnostic
comparisons, there were 51 children with DSM-IV ADI-II), and 41 control children, and
34 children with subthreshold ADHD symptomatology. One child with ADHD
hyperactive subtype was excluded from diagnostic comparisons. The 51 children with
ADHD included: 38 children with ADHD-Combined subtype (ADHD-C; 30% of total
sample), 12 children with inattentive subtype (ADD; 9%). The study focused on mothers
of children with ADHD. Data was collected from 51 fathers, thus there were more
children in the study than mothers as participation from both parents was not required.
Recruitment of teachers was successful (n=114); teacher data was used for diagnostic
purposes. Families were recruited through local school district mailings and from radio
and newspaper advertisements in the community. Of the families in the study, 70
families also participated in a larger study on neuropsychological and familial markers of
child ADHD. Thus, 55 families were recruited separately for purposes of this study. Just
over half of those 55 families (27/55) were screened by the smaller study and may or may
not have gone on to participate in the larger study after participation in the smaller study.
Recruitment sources were the same but the latter families completed a shorter battery of

measures and tests, given that they did not participate in Dr. Ni gg’s larger study.

52

Table 1 Child sample characteristics by diagnostic group

 

 

 

 

 

 

Total Control ADHD-C ADHD- ADI-ID F -test or
Sample inattentive subthreshold Chi-sq
P value
Sample size 126 41 38 12 34
Boys/girls 55% 54% 74% 42% 41% us.
70/58 22/19 28/10 5/7 14/20
(P=.05)
Age 9.6 9.8 9.58 (1.2) 9.3 9 us.
(1.5) 41.6) (1.5) (1.7)
Oppositional 35% 5% 66% 25% 39% as
Defiant Disorder 45/83 2/39 25/13 3/9 15/19
(% positive)
(ODD/no ODD)
Reading 10% 10% 8% 17% 11% us.
Disorder 13/128 4/37 3/35 0/12 6/28
% positive
(yes/no)

 

 

 

 

 

 

 

 

 

Notes: 126 is total sample; for primary analyses n=125, number of mother participants)
For sex: 1=male, 2=female
Procedure

Parents completed a battery of self-report and child-rating questionnaires. In 114
cases (89%), the child’s teacher also completed ratings of child behavior. Verbal consent
was obtained over the phone, and mothers completed a brief set of child ratings over the
telephone to screen families in or out of potential participation. If the family was
screened in, parents were mailed a copy of the full study consent form (later discussed in
a face to face meeting) along with a packet of questionnaires. The mother and target
child (child within the specified age range) came to campus. Maternal consent and child
assent were then obtained during a face-to-face meeting. The battery of questionnaires
and tests were then administered. The battery of questionnaires administered through the
mail and campus visit provided measures of three categories of variables: child
behavioral adjustment, parent attributions, and maternal role-specific adjustment. Child

attributions were also obtained for secondary analyses.

53

Measures
Child Behavior
Inattention-hyperactivity.
The ADHD Rating Scale-IV (DuPaul, Power, Anastopoulous, & Reid, 1998) was
designed to capture both Inattentive and Hyperactive symptoms of ADHD based on the
DSM—IV criteria. Respondents rated child behaviors on a four point Likert scale (“never

9’ 6‘

or rarely, sometimes,” “often,” or “very often”). Example items rated include: “fails to
give close attention to details or makes careless mistakes in schoolwork” and “Is ‘on the
go’ or acts as if ‘driven by a motor.” Each of the two subscales consists of nine items and
has satisfactory reliability and validity. Reported reliabilities are high (DuPaul, Power,
Anastopoulous, & Reid, 1998). For the parent form, reliabilities were: Inattention
alpha=.86 and Hyperactivity-Impulsivity alpha=.86. Reliabilities in the current study
(mother data) were: Inattention alpha=.94, and Hyperactivity-Impulsivity alpha=.93,
Total alpha=.96. In the current study, reliabilities for the parent form (father data) were:
Inattention alpha =.94, Hyperactivity-Impulsivity alpha =.91, and Total alpha l=.95. For
the teacher form, reported alpha reliabilities are: Inattention alpha =.96, Hyperactivity-
Irnpulsivity alpha =.88, and Total alpha =.94,. Teacher form in current study, reliabilities
were: Inattention alpha =.95, Hyperactivity-Impulsivity alpha =.95, Total alpha =.96.
Behavior Assessment System for Children

The Behavior Assessment System for Children (BASC; Reynolds & Kamphaus,

1992) Parent Rating Scales and Teacher Rating Scales contain Hyperactivity and

Attention Problems subscales. Items are rated on a 4-point scale, 0=never through

54

3=always. On the parent rating form, the Hyperactivity subscale consists of 10 iterrrs
such as “cannot wait to take turn,” and “leaves seat during meals.” On the parent rating
form, the Attention Problems subscale consists of 8 items such as “forgets things” and “is
easily distracted.” Reported reliabilities are high for both hyperactivity (alpha=.83), and
inattention (alpha=.77) (Reynolds & Kamphaus, 1992). Reliabilities in the current study
were adequate for mother data (hyperactivity alpha=.90; inattention alpha=.88) and father
data (hyperactivity alpha=.90; inattention alpha=.86). On the teacher rating form, the
Hyperactivity subscale consists of 13 items, such as “rushes through assigned work,”
“taps foot or pencil,” and “acts without thinking.” On the teacher rating form, the
Attention Problems subscale consists of 8 items, such as “has trouble concentrating” and
“is easily distracted fiom class work.” Reported reliabilities are high for both
hyperactivity (alpha=.92), and inattention (alpha=.87) (Reynolds & Kamphaus, 1992).
Reliabilities for teacher form within current study were good (alpha=.93 for hyperactivity
and .94 for inattention).
Conners ’ Scales-Revised—Short F arms

The Conners is another measure of disruptive behavior problems. For the parent
Conners, the manual test-retest reliability coefficients were .85 and .72, for the
Hyperactive subscale and (Conners, 1997). Internal reliabilities fiom our study for the
parent Conners’ were good for mother report (alpha= .92 for Hyperactivity subscale and
alpha=.96 for the ADHD Index) and father report (alpha=.92 for Hyperactivity subscale
and alpha=.93 for ADHD Index). For Teacher ratings, manual test-retest reliability

coefficients were .84 and .72, for these scales respectively (Conners, 1997). In the

55

current sample, internal reliabilities for the Teacher Conners hyperactivity and ADHD
scales were alpha = .92 and .95, respectively.
Child behavior: Oppositional/disobedient.

The Behavior Assessment Svste_m for Children (BASC; Reynolds & Kamphaus,
1992) parent teacher Aggression and Conduct Problem subscales were used to measure
child disruptive behaviors. The BASC parent rating form Aggression (alpha=.89),
Conduct Problems (alpha=.82) and the BASC teacher rating form Aggression (alpha=.94)
and Conduct Problems (alpha=.74) have established reliability and validity. Reliabilities
within the current study were satisfactory for mother ratings (Aggression alpha=.89,
Conduct Problems alpha=.84), father (Aggression alpha=.9l, Conduct Problems
alpha=.89), and teacher ratings (Aggression alpha=.94). The reliability of the Conduct
Problem subscale for teacher ratings was inadequate (alpha=.59) and it was not included
in subsequent analyses.

The Conners Rating Scale-Revised (Conners, 1997) Oppositional Behavior
Subscale behaviors subscale also provided a measure of child oppositional/disobedient
behavior. Parents and teachers rated children on a four point scale, 0=not true at all to
3=very much true. Example items include: “Angry and resent” and “argues with adults.”
Manual reported reliability for both parent (alpha=.92) and teacher forms (alpha=.88) are
satisfactory. Alphas were also high within current study (mother rating alpha=.92, father
rating alpha=.93, and teacher rating alpha=.94).

ADHD Diagnosis.
A multistage screening and diagnostic process was used in determining child

diagnosis.

56

At stage 1, children were» considered My]; ADHD in two ways. (a) They
exceeded screening cut-offs on at least one current parent and teacher rating scale:
Behavior Assessment Scale for Children (BASC, Reynolds & Kamphaus, 1992)
hyperactivity or inattention scale T > 60, Conners (1997) Rating Scale hyperactivity
index T > 60, or at least 4 symptoms of inattention or hyperactivity endorsed [with a
rating of “2" or “3"] on the DuPaul ADHD Rating Scale (DuPaul, Power, Anastopoulous,
& Reid, 1998). (b) They were previously diagnosed as ADHD by a physician or
psychologist, who included teacher and parent ratings to make the diagnosis. Children
were considered possible Controls if below cutoffs on all parent and teacher scales and
never diagnosed with ADHD in the community.

ADI-ID diagnostic assignment and subtype was then confirmed using an “or”
algorithm with the Diagnostic Interview Schedule for Child DSM-IV diagnosis (DISC-
IV). The “or” algorithm uses the parent Diagnostic Interview Schedule for DSM-IV
(DISC-IV, Shaffer, Fisher, & Lucas, 2000) for parent rating of symptoms supplemented
by available teacher ratings as described above. The DISC-IV is a structured diagnostic
interview developed by NIMH. It implements a stringent diagnostic algorithm requiring
onset before age 7, persistence of symptoms greater than 6 months, and symptom-specific
impairment in at least two settings. Prior versions of the DISC have exhibited acceptable
reliability and validity (Shaffer et al., 1993). The computer-assisted interview was
administered to the child’s primary caregiver by trained interviewers. A symptom was
counted as “present” if endorsed by mother a; teacher. Provided all other criteria were
met and they had at least four symptoms on the DISC-IV, those symptoms were added up

to determine ADHD status and subtype. Cases with five symptoms of inattention or

57

overactivity by this method were placed in the “subthreshold” group. According to field
trial data, they might have ADHD-C or ADHD-I (Lahey et al., 1994).
Parent Attributions

Parental attributions about child behavior were assessed using two primary
measures: the Written Analogue Questionnaire and the Recalled Incident Interview. For
exploratory analyses, an additional measure was used, namely Hoza’s Interactions
Questionnaire. All three measures were designed specifically for studies of attributions
about disruptive child behaviors. The first two measures separate attributions about child
Inattentive-Overactive behaviors from child Disobedient/Non-compliant behaviors, and
for this reason were used as primary measures for the current study. [Again, it is noted
that attributions were about child’s not mother’s behaviors].

Written Analogue Questionnaire.

The Written Analogue Questionnaire (W AQ; see Johnston & Freeman, 1997) is a
self-report questionnaire where parents rate 12 standard, hypothetical but common
behavioral scenarios (4 for each of three behavior types: 1) inattentive-overactive, 2)
oppositional- defiant, and 3) prosocial-positive). Before completing the set of questions,
parents were given an explanation the attribution dimensions. After reading each
scenario, parents rated on lO-point scales the (1) causal locus, (2) controllability, (3)
stability, and (4) globality; (5) parent responsibility for the behavior and (6) affective and
(7) behavioral responses which the parent might engage in following such a behavior by
their child. In support of reliability and validity, Johnston (Johnston & Freeman, 1997)
estimated internal consistencies by correlating responses to two examples of each

behavior type. Correlations ranged from .10 to .82, with median correlation of .49.

58

Within the current study, analyses focused on attribution ratings of locus, controllability,
stability, and globality. There were four items for each of the above attribution rating
and per behavioral type (inattentive-overactive, oppositional/disobedient,
prosocial/positive). An example scenario follows: “Your child enters the kitchen as you
have finished sweeping the floor and getting the dirt in a pile to pick up. He/she doesn’t
wait for you to finish and heads straight to the fiidge. As he/she rushes through the
kitchen, the pile of dirt scatters across the floor.” In the current study, alpha reliability
coefficients were obtained and were adequate (coefficients ranged fiom .70 - .87, except
for measures of locus which were lower than .70.
Recalled Incident Interview

Johnston’s (see Johnston & Freeman, 1997) semi-structured interview was used to
obtain parent attributional ratings of parent reported, actual child behavioral incidents
recalled by parents. For this measure, the parent respondent was asked to recall two
recent specific instances of irnpulsive-overactive, oppositional, and prosocial behaviors.
A senior research assistant interview administered the interview following a standardized
protocol whereby causal attributions were explained and example behaviors were
provided. The parent was redirected to the example behaviors if he/she provided a
behavior which was not clearly either impulsive-overactive, oppositional, or prosocial.
After describing each recalled behavior, the parent rated (1) the intensity of the behavior;
(2) causal locus, (3) controllability, (4) stability, (5) globality/generalizability, (6) parent
responsibility for the behavior, parental (7) affective and (8) behavioral responses. In
support of reliability and validity, Johnston (Johnston & Freeman, 1997) estimated

internal consistencies by correlating responses to two examples of each behavior type.

59

Correlations ranged from .03 to .58, with median correlation of .32. Within the current
study, analyses focused on attribution ratings of locus, controllability, stability, and
globality. Two recalled incidents were obtained for each of three behavior types:
inattentive-overactive, oppositional/disobedient, and prosocial/positive. Reliability
ranged from .33 - .70, notably variable.
The Interactions Questionnaire

The Interactions Questionnaire (Hoza & Pelhm, 1995) was used for the testing of
an exploratory hypothesis (exploratory hypothesis 2). The Interactions Questionnaire
was developed by Hoza and Pelham in order to assess parents’ attributions pertaining to
their children’s compliance and noncompliance. The measure consists of 6 brief
hypothetical but common scenarios of child non-compliance or compliance each
followed by ten questions parents rate on a 10—point scale (l=really true to 10=not true at
all). Note that unlike the Written Analogue Questionnaire (W AQ) and Recalled Incident
Interview (RH) which separate Inattentive-Overactive from OppOsitional/Disobedient
child behaviors, within the Hoza Interaction Questionnaire behaviors are either
compliance or non-compliance. In the original Hoza Interactions measure, attribution
dimensions were assessed and scored separately for compliance and noncompliance.
Three compliance and three non-compliance scenarios were administered; however, due
to time constraints in the current study, data were only collected for 6 attribution
dimensions. That is, parents rated each of the 6 scenarios (3 per each behavior type) on
the six most relevant attribution dimensions rather than on the 10 attribution dimensions
assessed in the complete, original Hoza Interactions measure. Thus, in the current study,

parents rated the extent to which child behavior was due to: (1) child mood, (2) child

60

effort to please parent, (3) child ability to control him/herself, (4) parent effort to control
child’s behavior, (5) parent general quality as parent (“good parent”) and (6) parent
ability to obtain child behavioral compliance. These subscales were selected due to
their similarity to attribution dimensions reviewed and targeted for study. Hoza et al.
(2000) reported coefficient alphas for the lack of effort (alpha=.87) and child mood
(alpha=.85) subscales.

Although six attribution dimensions were included in the data collection effort,
only three of the attribution subscales were proposed for use: (1) parent effort to control
child’s behavior, (2) parent general quality as parent (“good parent”) and (3) parent
ability to obtain child behavioral compliance. These subscales were used as attempt to
test exploratory hypothesis 2. Reliabilities for the first two subscales were low
(alpha=.60 and .58), corresponding to the factor analytic results presented in the section
on data reduction. These two subscales were not used for testing main hypotheses. The

third subscale had adequate reliability (alpha=.90).

Parent Role Adjustment
Satisfaction with Parenting Performance Scale
The Satisfaction with Parenting Performance subscale of the Parenting
Satisfaction Scale (Guidubaldi & Cleminshaw, 1994) contains fifteen items, (rated on a
4-point scale, l=Strongly Agree, 2=Agree, 3=Disagree, 4=Strongly Disagree). Items
include: “I wish I did not become impatient so quickly with my child” and “I wish I were
a better parent and could do a better job of parenting.” (Note: these two items are

reversed scored so that higher score equates more satisfaction). The internal consistency

61

reliability reported in the manual is satisfactory (F82). Within the current sample,
reliability was also satisfactory (alpha=.85 for maternal self-ratings, alpha=.86 for
paternal self-ratings).
The Parenting Stress Index- Short F arm

The Parenting Stress Index — Short Form (PSI; Abidin, 1995) consists of 36 items,
to be rated on a 5-point scale (strongly agree to strongly disagree). Example items
include: "I often have the feeling that I cannot handle things well,” "I find myself giving
up more of my life to meet my children's needs than I ever expected."

The PSI Total Stress score was chosen for use in the current study. Reliabilities
for the PSI Total Stress score within current sample were satisfactory (alpha=.93 for
mother self-rating, alpha=.90 for father self-ratings). Additional information is included

in section on data reduction.

Measure of Child Attributions
Written analogue questionnaire.

A child version of Johnston’s Written Analogue Questionnaire was adapted by the
author from the parent version described below. Adaptations were aimed at simplifying
language and scaling complexity so that it could be understood by children as young as 7
years old. The format was broken into two parts where children first made a
dichotomous decisions about each attribution (i.e., “something about you” or “something
not about you”) and then were asked to rate dimensionally (i.e., “A lot about you”,
“Mostly about you”, “some about you”). A picture accompanied the scenarios that were

read in an engaging way to the child. In order to keep the child engaged in the task, three

62

specific items were personalized by asking the child specific questions and inserting his
or her response into the behavioral scenario. For example, for an item involving a child
looking for his/her sports equipment, the child was first asked what if any sports, he she
played. That sport with appropriate equipment was then inserted into the scenario.
Children were asked to imagine that they had engaged in the behavior and then rate the
attribution dimensions. The written analogue questionnaire was administered following a
standardized protocol. Because the current author created the child version as an
adaptation of Johnston’s adult measure, there is no prior reliability information on this

measure. Notably, this is an exploratory, pilot sub-study.

Additional Child Variables
Reading Disability (RD).

The Wechsler Intelligence Scales for Children-III (WISC-IH; Wechsler, 1991)
short form and Wechsler Individual Achievement Test (WIAT; Wechsler, 1992) screener
were administered. Two criteria were required for meeting criteria for a Reading
Disorder: (1) standardized reading score on WIAT reading < 85 and (2) IQ (ability) -

achievement discrepancy equal to or greater than one standard deviation (15 points).

ODD or C D Diagnoses.
Diagnosis of Oppositional Defiant Disorder and diagnosis of Conduct Disorder
were calculated from the Diagnostic Interview Schedule for Children (DISC-IV). If
children met criteria based on mothers’ DISC interview, they were assigned a diagnosis

of ODD or CD as appropriate. Note that the DISC interview has parents rate each of the

63

behaviors under ODD and CD per the DSM-IV as well as onset, severity, and impairment
into functioning so that the interview provided the ability to make a diagnoses based on

the DSM-IV criteria for ODD and CD.

Data Reduction

Multiple instruments were used to measure various constructs. In order to
simplify data and most reliably measure the intended constructs, composite scores were
created when justified. Factor analyses were conducted in order to confirm factor
structures for measures that have been relatively newly developed (i.e., attribution
measures). Inter-correlations were conducted when appropriate in order to further test
whether composites were warranted/appropriate/justified.

Measures pertaining to ratings of child behavior are discussed first, followed by
maternal attribution measures, maternal role adjustment measures, and child attribution
measures. Because the study focus is on maternal attributions and adjustment, results for
mothers are presented first (throughout remainder of document). Child attribution results
are most exploratory and are presented last. Teacher data (n=114) was used for

additional testing of results when appropriate.

Data Reduction of Adult Reported Ratings of Child Behavior
Within-reporter composites were created for the three core child behavior
domains (inattention, hyperactivity, and disruptive behavior). Teacher and mother
ratings were not combined for dimensional analyses. Maternal ratings of child behaviors

may certainly be viewed as measure of maternal —percaption or experience of severity of

64

child behavior problems. Interpretations are limited by this constraint as discussed later.
When relevant, results were cross-checked with teacher ratings which may be viewed as a
cross validation of results as well as a check on cross situation extent of observed
association.

For mother, father, and teacher ratings of child behavior, confirmatory factor
analyses were conducted separately for each rater in order to confirm a three-factor
solution (inattention, hyperactivity, opposition/aggression) for each rating of child
behaviors. The three-factor confirmatory analyses resulted in a factor for child (a)
inattention, (b) hyperactivity and (0) other disruptive behaviors. The three-factor
solutions were confirmed for each rater. As the measures have established, normed
factors, principal components confirmation factor analyses was conducted with the
established factors to validate the use of three child behavior factors. Unsurprisingly, the
three components were confirmed across measure as expected (that is into the three
components noted above) with eigenvalues all greater than 1.

For each rater, a composite score was then created for each component factor.
The Inattention factor was created with the average ratings on ADHD Rating Scale
inattention subscale and BASC Inattention scale. The hyperactivity factor was created by
averaging the scores on the ADHD Rating Scale Hyperactivity subscale, Conners’
Hyperactivity scale, and BASC Hyperactivity Scale. The disruptive Disobedient/
Oppositional behaviors factor consisted of the average of the rater’s ratings on the
Conners’ Oppositional subscale, BASC Conduct Problems subscale, and BASC
Aggression subscale. Each of the composite scores were created for each rater (mother,

father, teacher) and yielded adequate reliabilities.

65

Composites for Maternal Ratings of Child Behaviors.

Composites of mother ratings across measures yielded composites with good
internal reliabilities (inattention alpha=.91, hyperactive alpha=.95, and oppositional
alpha=.90,). As expected from the literature, oppositional defiant/aggressive behavior
and hyperactivity were highly correlated in the sample (F83, p<.001; see also Table 2).

Table 2: Correlations between three-factor solution for mother ratings, (reliabilities on
diagonal)

 

 

Inattention Hyperactivity ODD
Inattention .91
Hyperactivity .69""' .95
ODD 57“" .83‘" .90

 

mp<.ool

The second step in data reduction was to look at inter-correlations among
measures again to confirm that composites were justified. Within rater, inter-correlations
supported the creation of composites by behavior type. For each behavior type
(inattention, hyperactivity, and oppositional/defiance) correlations were greater than .80
(all p<.001; see Table 3). Notably, cross construct correlations were also high (ranging
from .49-.69) but somewhat less than .80. Thus, the different behavior domains are
arguably partially separate although highly correlated. These correlations along with the
principal components confirmatory factor analyses supported the retention of the three

behavioral factors, albeit with a caution as to their overlap.

66

Table 3: Inter-Correlations among child behavior measure subscales — mother ratin

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

ARS ARS BASC BASC BASC BASC Conners Conners Conners
lnattentio Hyperacti lnsttentio Hyperacti Aggressio Conduct Oppositio Hyperacti ADHD
n vity n vity n Problems nal ve Index

ARS 1.0

Inattentio

n

ARS .73‘" 1.0

Hyperacti

vigy

BASC .88‘“ 61"" 1.0

lnattentio

n

BASC 453""I .88‘” .59‘“ 1.0

Hyperacti

Vity

BASC .55‘“ 74"” .51‘” .81‘“ 1.0

Aggressio

n

BASC .49‘" .66‘“ .48‘" .71‘“ .78‘“ 1.0

Conduct

Conners .65‘“ .78“. .60‘" .80‘" .82‘” .75‘" 1.0

Oppositio

rial

Conners .69'“ .91‘” .62‘” .88‘” .75'“ .67‘“ 83"“ 1.0

Hyperacti

ve

Conners 92"" .76‘“ .88‘" 10‘” .61‘” .55‘" .69‘” .77‘” 1.0

ADHD

Index

ARS-ADHD rating scale

BASC-Behavioral Rating Scale for Children
”'p<.001, “p<01, ’p<.05, +p<0.l

Teachers ’ Ratings of Child Behaviors.

Confirmatory factor analysis was used to confirm a three-factor solution for

teacher data. As with mother and father data, factor composites were created for

Hyperactivity, Inattention, and Disruptive behaviors. Reliabilities were satisfactory for

all three composites (See Table 4). The reliabilities for teacher ratings of child

Oppositional/Disobedient (alpha=.9l) and hyperactivity (alpha=.93) were high and

reliability for composite teacher rating of child inattention (alpha=.88) was satisfactory.

67

 

Table 4 Correlations for teacher behavior rating composites, 3-factor solution
(reliabilities on diagonal)

 

Inattention Hyperactivity ODD
Inattention .88
Hyperactivity .79"* .93
ODD .77‘" 86“" .91

 

mp<.001

Correlations Between Maternal and Teacher Data.
The correlations between mother and teacher composite ratings of child behavior
are shown here (Table 5). Mother and teacher ratings were highly correlated.

Table 5 Correlations between mom and teacher ratings for three-factor solution

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Mom rated Mom rated Mom Teacher Teacher Teacher
Inattn Hyp rated ODD rated Inattn Rated Rated
Hyp ODD
Morn rated Inattn
1.0
Mom rated Hyp
.69‘” 1.0
Mom rated ODD
.57‘" 83"" 1.0
Teacher rated Inattn
70"" .48"" .38’" 1.0
Teacher Rated Hyp
52"" .62‘“ .54‘" .65‘" 1.0
Teacher Rated ODD
. 38"" .58‘" .61"" .45‘" .73‘“I 1.0
"“p<.001

Data Reduction of Maternal Attribution Data
For each of Johnston’s measures (Written Anaologue Questionnaire, WAQ, and
the Recalled Incident Interview, RII), preliminary analyses were conducted to determine
factor structure. The steps were identical for each measure; however, for clarity, each
measure is discussed separately. For each measure, data reduction was conducted first
for mother data and then father data. Because these attribution measures are not yet well
established or validated, exploratory factor analyses used run to guide the creation of

subscales.

68

 

Written Analogue Questionnaire (WAQ)

Factor analyses were conducted in order to determine the factor structure within
current dataset. Within the WAQ, three behavior types were rated (Inattentive-
Overactive, Disobedient/Oppositional, and Prosocial/Positive). Four scenarios of each
behavior type were administered. Finally, four primary attribution domains were
included: locus, controllability, stability and globality. Thus, for each behavior type (3),
there were four scenarios (4) and four attribution ratings (4), resulting in 16 items per
behavior type (48 items total)3. Oblimin rotation with Kaiser Normalization was used.

The literature suggests that attributions differ for different types of behaviors,
such as positive versus negative. Thus, factor analyses were conducted separately by
behavior type.

The factor solutions by behavior type are shown in Tables 6, 7, and 8. As shown
in Table 6, a four-factor solution emerged for attributions about Inattentive-Overactive
behaviors. Items with cross-loadings greater than .25 were deleted in order to obtain a

clean factor solution". (Dropped items are not shown).

 

3 Note: a few additional items were asked about each behavior type and scenario; however, those results are
presented after considering the reduction of the four attribution domains of primary interest in the current
study.

’ Defined as factor solution with minimal cross-loadings greater than .25.

69

Table 6 Factor loadings of Maternal Attributions for Inattentive /Overactive behaviors
via WAQ Questionnaire Scenarios

 

 

Factors Individual Variable Labels Factor 1 Factor 2 Factor 3 Factor 4
Globality (Eigenvalue=3 .23)
Scenario a globality .80 -.10 .19
Scenario d globality .77
Scenario c globality .71 -. 14
Scenario b globality .67 .13
Locus (Eigenvalue=2.6)
Scenario a locus -.11 ' .89 -.20
Scenario d locus .71 .22
Scenario b locus .20 .62 .16
Controllability (Eigenvalue=l .2)
Scenario c controllability -.ll .90
Scenario d controllability .13 .80 -. 14
Stability (Eigenvalue=l .2)
Scenario c stability .84
Scenario a stability .12 .79
Scenario d stability -. 15 .70

 

For Oppositional/Disobedient behaviors, no clean factor solutions emerged until
the globality factor was dropped. After dropping items with multiple cross-loadings,

stability and globality loaded as one factor (see Table 7).

 

5 Defined as factor solution with minimal cross-loadings greater than .25.

70

Table 7 Factor Solution of Maternal Attributions for Oppositional / Disobedient
behaviors via WAQ Questionnaire Scenarios

Factors Individual Variable Labels Factor 1 Factor 2 Factor 3
Stability/Globality (Eigenvalue=3.3)

Scenario h stability .75 -.32

Scenario f stability .73

Scenario g globality .69

Scenario h globality .67

Scenario e stability .66 -.3O

Scenario f globality .63 .26

Locus (Eigenvalue=2.6) .83
Scenario e locus .72
Scenario f locus .65
Scenario g locus

Controllability (Eigenvalue=1.2)
Scenario e controllability .85
Scenario h controllability .79
Scenario f controllability .76

 

Notably, globality and stability have been viewed as one factor in prior studies
with other measures of attributions (Bunce & Peterson, 1997; Peterson, Seligman, Yurko,
Martin, & Friedman, 1998) as well as by Geller and Johnston (1995) on the WAQ and
RH. However, because globality and stability emerged as distinct factors for the
Inattentive-Overactive and Prosocial/Positive behavior types, the decision was made to
drop globability for Oppositional/Disobedient behavior type rather than combining it with
stability. Thus, rather than combining globality and stability for the WAQ OPP/D
(Oppositional/Disobedient) attributions, globality was dropped. The final factor solution

with remaining cross-loadings is shown in Table 8.

71

Table 8 Final Factor Solution of Maternal Attributions for Oppositional / Disobedient
behaviors via WAQ Questionnaire Scenarios (globality items dropped)

 

Factors Individual Variable Labels Factor 1 Factor 2 Factor 3
Controllability
Scenario f controllability .81
Scenario e controllability .81
Scenario h controllability .75
Scemrio g controllability .69 .12
Stability
Scenario f stability .85
Scenario h stability .81 .21
Scenario e stability .80
Scenario g stability -.1 l .77
Locus
Scenario e locus -.80
Scenario g locus .11 -.68
Scenario f locus .20 -.67
Scenario h locus -. 15 -.60

 

For mother attributions about Positive behaviors, a clean four-factor solution
emerged as shown in Table 8; only one item was dropped due to cross-loadings greater

than .25.

72

Table 9 Factor loadings of Mom Attributions for Positive/Prosocial Behaviors via WAQ
Questionnaire Scenarios

 

Factors Individual Factor 1 Factor 2 Factor 3 Factor 4
Variable labels
Stability
Scenario j 86
Scenario 1: .85
Scenario 1 .83
Scenarioi 64 .17 -.15
Controllability
Scenario 1 .87 . 15
Scenario j .84 .16
Scenario i -.15 .81 -.18
Scenario k .79
Locus
Scenario k .11 .77
Scenario j .13 .71
Scenarioi -.15 .70
Globality
Scenario k .15 -.86
Scenario j -.83
Scenario 1 .17 -.80
Scenario i . 13 .25 -.14 -.67

 

Creation of Subscales and Reliabilities.

Subscales were created by averaging each rating of attribution domain across
scenarios. Subscales were created for Inattentive-Overactive locus, control, stability, and
globality; Oppositional/Disobedient locus, control, and stability, and Positive behavior
locus, control, stability, and globality.

As shown in Table 10, the WAQ attribution subscales generally had satisfactory
reliabilities. Locus had a lower than preferred reliability across behavior domains
(alpha=.66 for 10 and Oppositional behavior type, alpha=.65 for positive behavior type).
All other reliabilities were greater than .70, which are still lower than desired but more

adequate.

73

Table 10 Intercorrelations among Mother rated WAQ factors and data reliabilities (on

 

 

 

 

 

 

 

 

 

 

 

 

 

diagonal) of composrtes
10 10 IO 10 OPP/D OPP/D OPP/D PRO Positive Positive Positive
Locus control stability globality locus control stablity lPositive control stability globality
locus

10 .66

Locus

IO .24" .75

control

10 .00 -.28“ .70

stability

10 .24“ -.13 .4?“ .75
MW

OPP/D .47'” .31‘" -.08 .12 .66

locus

OPP/D .19" .69‘" -.29"'" -.05 .42‘" .77

control

OPP/D -.10 -.27” .68‘" .34‘" -.09 -.28" .83

stablity '

Positive .17 .06 -.1 l -.02 .30‘" .l 1 -.10 .65

locus

Positive .05 .47‘" -.20" -.03 .25" .55‘" -.21‘I .28” .87

Control

Positive .01 .23' -.34‘” -.29“" .09 .21“ 45‘“ .27“ 35"“ .84
Stability

Positive .08 .25" -.43”“ -.23’ .17+ .30‘" ~56’" .39‘” .35‘“ .55‘” .86
Globality

 

 

 

 

 

 

 

 

 

 

 

 

mp<.oor, "p<.01, *p<.05, +p<o.1

Inter-correlations revealed that type of attribution dimensions (i.e., locus, control,
stability, globality) was significantly related for Inattentive/Overactive and
Oppositional/Disobedient behaviors and that these correlations were medium to large in
magnitude. This supported the validity of the scales.

In summary, the resulting factors for mother WAQ attributional domains for
testing of hypotheses are noted herein. For Inattentive-Overactive (IO) and
Positive/Prosocial (PRO), the result was a four-factor solution: (1) locus, (2)
controllability, (3) stability, and (4) globality. For Oppositional/Disobedient behavior
type, globality dropped and a three-factor solution: (1) locus, (2) controllability, and (3)
stability were used. Reliabilities were adequate (alpha>.70) for most subscales (except
for locus, alpha<.70). Reliabilities for Oppositional/Disobedient stability rating and
Prosocial/Positive globality, stability, and controllability were within desirable range

(alpha>.80).

74

 

Recalled Incident Interview (RH) Data Reduction.

As with the WAQ, an initial factor analyses with all items included did not reveal
usable factors. The factor structure was not clean, with multiple overlapping items
correlating greater than .25 (as well as no clear break in the scree plot; KMO Measure of
Sampling Adequacy=.66). Therefore, as was done with the WAQ, factor analyses were
conducted by behavior domain.

For the recalled incident interview, only two recalled incidents were recalled per
behavior type (in contrast to four scenarios per behavior type in the WAQ). Thus, for
each behavior type (3), there were two recalled incidents (2) and the four attribution
domains (4) under consideration, resulting in 8 items per behavior type (total of 24
items).

For Inattentive-Overactive behavior type, a clean factor solution emerged during

the first run (see Table '11).

75

Table 11 Factor Solution of Maternal Attributions for Inattentive/Overactive behaviors
via RII Questionnaire Scenarios
Factors Individual Variable Labels Factor 1 Factor 2 Factor 3 Factor 4

 

Stability (Eigenvalue=2.2)
Recalled Behavior 1 stability .90
Behavior 2 stability .84 -.10

Locus (Eigenvalue=l .5)
Recalled Behavior 2 globality .88
Behavior 1 globality .12 .78 .20

Controllability (Eigenvalue=l .2)
Recalled Behavior 1 controllability .88
Behavior 2 controllability .87

Globality (Eigenvalue=1.0)
Recalled Behavior 2 locus -. l 8 -.90
Behavior 1 locus .41 -.57

 

For Oppositional/Disobedient behavior type, the initial factor analyses resulted in
a 3-factor solution with globality items splitting and collapsing with stability and

controllability (See Table 12).

76

Table 12 Factor Solution of Maternal Attributions for Oppositional/Disobedient

behaviors via R11 Questionnaire Scenarios

 

Factors Individual Variable Labels Factor 1 Factor 2 Factor 3

Stability/Globality (eigenvalue=2.3)

Recalled Behavior 2 stability .84 .ll
Behavior 1 stability .77 -.22
Behavior 2 globality .56 .45 -.22

Locus (eigenvalue=l .5)

Recalled Behavior 2 locus -.23 .71 -.12

Behavior 1 locus -.23 .66 .30

Controllability (eigenvalue=l .1)

Recalled Behavior 1 controllability .ll .88
Behavior 1 globality .21 .43 -.55
Behavior 2 controllability .30 .32 .45

 

For consistency with other literature and behavior types, globality items were

dropped, resulting in a 3-factor solution with subscales in acceptable factor structure and

ties to literature. Notably, when globablity items were dropped (as was done with the

WAQ), a clean 3-factor solution remained (see Table 13).

Table 13 Factor Solution of Maternal Attributions for Oppositional/Disobedient
behaviors via RII Questionnaire Scenarios (globality items dropped)

 

 

Factors Individual Variable Labels Factor 1 Factor 2 Factor 3

Stability (eigenvalue=l .8)

Recalled Behavior 2 stability -.85 .20
Behavior 1 stability -.81 -.19

Locus (eigenvalue=l .3)

Recalled Behavior 2 locus .16 .84 -. l 7

Behavior 1 locus -.22 .73 .20

Controllability (eigenvalue=l .1)

Recalled Behavior 1 controllability .89
Behavior 2 controllability .24 .68

 

For positive/prosocial behaviors, initial analysis resulted in a three-factor solution;

however, rather than attribution dimensions emerging as factors, three of the attribution

domains (locus, stability, and globality) clustered by behavioral example (recalled

behavior 1 and recalled behavior 2) rather than by attribution (see Table 14). That is,

those three attribution domains did not separate but rather clustered together for each
recalled behavioral incident. Only for the attribution domain of controllability did
controllability items cluster together. Therefore, only the controllability attribution
domain was used for analyses when concerning Recalled Incident Interview Prosocial
Behaviors. Factor analytic results for Recalled Incident Prosocial Behaviors are shown in
Table 14.

Table 14 Factor Solution of Maternal Attributions for Positive/Prosocial behaviors via

RII Questionnaire Scenarios
Factors Individual Variable Labels Factor 1 Factor 2 Factor 3

 

Factor 1 (eigenvalue=2.4)

Recalled Behavior number 2 stability .85 -. 10
Behavior 2 locus .72
Behavior 2 globality .70 .13

Factor 2 (eigenvalue=l .5)

Recalled Behavior number 1 globality .78
Behavior 1 stability .75
Behavior 1 locus .69

Factor 3 (eigenvalue=1.0)
Controllability
Scenario 1 controllability -. 14 ' .ll .90

Scenario 2 controllability .24 -.14 .65

 

For completeness, inter-correlations among the attribution subscales are presented
in Table 15 along with reliabilities. The subscales for the RH generally had poor
reliabilities (alphas<.70). As with the WAQ, attributions about locus generally had
lowest reliabilities (alpha=.33 for 10 locus, alpha=.42 for OPP locus). IO stability had

marginally adequate reliability (alpha=.70).

78

Table 15 Reliabilities of (on diagonal) and intercorrelations among mother rated RII

 

 

 

 

 

 

 

 

 

factor composites
10 IO IO 10 OPP/Dis OPP/D OPP/D POS/Pro
locus control stability lobality locus control stability control

10 locus .33

10 control -.09 .68

IO stability .13 -.18"' .70

IO globality 32"" -.02 .12 .65

OPP/D locus .28“ .09 .06 29*" .42

OPP/D -.07 .49”* -.18" -. 17+ .03 .5 1

control

OPP/D -.04 -.04 .49*** .17 -.07 -.28*"‘ .64

stability

POS/Pro -.09 .24" -.04 .00 .08 .42“" .04 .45

control

 

 

 

 

 

 

 

 

 

 

 

mp<.oor, "p<.01, ‘p<.05, +p<o.1

Surprisingly, 10 Stability and Globality were not highly related, nor were
controllability and locus (for 10 or OPP behaviors).

Attribution ratings which were highly correlated included: 10 control and OPP
control (r=.49, p<.001) and 10 stability and OPP stability (r=.49, p<.001) suggesting that
control and stability attributions are similar for 10 and OPP ratings but that the 3 or 4
factor solution better fit the current data compared to the 2 factor solution used in some of
Johnston’s prior work (Geller & Johnston, 1995).

In summary, the Recalled Incident Interview factors were less clean compared to
factors resulting from the Written Analogue Questionnaire. For Inattentive-Overactive
behavior type, a clean four-factor solution emerged, allowing creation of scores for each
type of attribution (locus, controllability, stability, and globality). For
Oppositional/Disobedient behavior type, globality did not emerge as a useable factor so
factors are available only for locus, controllability, and stability. For the

Positive/Prosocial behavior type, the factors were grouped largely by the recalled

79

behavioral incident given by mothers; the only factor that emerged for attribution domain

was controllability.

The reliabilities of the R11 subscales were much lower when compared to the

reliabilities obtained with the WAQ. The RII reliabilities were likely reduced by the

fewer number of items (two opposed to four used in WAQ) as well as by the greater

variability per scenario. Recall that in the RI], parents were asked to give an example

behavior in which the child had actually engaged. The severity of the behaviors varied

considerably as some of the children exhibited symptoms of IO significant enough for a

diagnosis of ADHD whereas others likely exhibited these behaviors within average or

“non-disordered” level of severity, typical of most children during this stage of

development. Thus, the lower reliability may be accounted for by the structure of the

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

measure.
Table 16 Correlations between WA and RH subscales
WAQ WAQ WAQ WAQ WAQ WAQ WAQ WAQ WAQ WAQ WAQ
IO [0 IO 10 OPP/D OPP/D OPP/D Pro Pro Pro Pro
locus control stability globality locus control stability locus control stability globality
R11 10 .21 " -.01 .01 . 16+ .04 -.06 -.07 .04 -.06 -.07 -.
locus
RI] IO .07 .56‘“ -.15 -.06 .18‘ .48‘” -.10 .18‘ .48‘” -.10 .02
control
R11 10 -.12 0.10 .50‘“ .21‘ .004 -.14 .49‘” .004 -.14 .49‘” .21‘
stability
R11 10 .08 -.06 .26" .43‘“ .02 -.02 33"" .02 -.01 .33‘" .41“.
globality
R11 .25" .05 .l l .l l .09 .03 .06 .09 .03 .06 .06
OPP/D
locus
R11 .12 .5l‘” -.23“ -.01 .25” 52"” -.12 .25” .52‘" -.12 -.05
OPP/D
control
RI] -.1 1 -.Ol .46‘" .25“I -.06 -.14 .47‘" -.06 -.14 .47"‘ 34’”
OPP/D
stability
RII Pro .002 32"“ -.03 -.03 .l9’ .29" -.03 .20" .29” -.03 .05
control
mp<.oo1, "p<.01, ‘p<.05, +p<.1

80

 

Data Reduction for Measures of Maternal Role Adjustment
Parent Satisfaction Survey and Parenting Stress Index Scales.

The Parenting Satisfaction Survey Satisfaction with Parenting Performance was
considered as an outcome variable as was the PSI Parent Distress Subscale. Because
normative data is available for the PSI-SF Total score, it was considered as a possible
outcome in place of the PSI Parent Distress subscale.

As shown in Table 17, reliabilities within current study for parent role adjustment
(used as outcome measure) were satisfactory. The reliability for the PSI-SF Total Stress
score (alpha-.93) was expectedly (due to larger number of items used in total score)
higher than the reliability for the PSI Parent Distress form (alpha=.85).

Table 17 Reliabilities and inter-scale correlation matrix for maternal role adjustment

 

 

 

 

 

 

 

 

 

 

PSS Satisfaction with PSI Parent Distress PSI-SF Total Stress
parentirigperformance
PSS Satisfaction with .85
parentingpgrformance
PSI Parent Distress -.48*“ .85
PSI-SF Total Stress -.50”"' 79"" .93
‘*"'p<.001

Both the PSI-SF Total Stress Score and PS1 Parent Distress score were similarly
related to PSS Satisfaction with Parenting Performance (r= -.48 and r= -.50,
respectively), they were considered comparable usefulness in current study. Due to its
usefulness for comparison with the literature, the PSI-SF Total Stress score was used
instead of the PSI Parent Distress score.

Thus, the PSI-SF Total Stress Score and PSS Satisfaction with parenting

performance were retained for analysis.

81

Additional Attribution Measures
Attribution Items Pertaining to Parent Responsibility and Control

Additional Attribution Items of Interest (needed for Exploratory Analysis _2:)_ In
prior studies, attributions about child controllability have been found as key to parental
adjustment (Dix & Grusec, 1985), the extent to which parents feel responsible for their
child’s behavior and the extent to which they feel both responsible and able (or unable) to
control their child’s behavior might be important in their adjustment. An additional
exploratory analysis is that parental experience of responsibility might interact with
perceived ability to control child’s behavior might predict parental role adjustment.

Two items were added to the WAQ in order to assess the extent to which parents
reported feeling responsible and able or unable to control their child’s behavior. Factor
analysis of these two items (asked four times per each of the three behavior types, except
Prosocial for which only parental responsibility was asked) revealed that they form as
two separate items. The factor analytic results are shown in Tables 18 and 19. Again,
the bend in the scree plot and eigenvalues greater than 1 along with correspondence to

literature were used to guide final factor solution selection.

 

6 Note: parent exploratory analyses conducted for mother data only

82

Table 18 Maternal WAQ Attributions about parent role in child behavior factors for

Inattentive-Overactive Behaviors
Factors Individual Variable Labels Factor 1 Factor 2

 

Parental ability to control child behavior
(eigenvalue=2.9)
Scenario 3 controllability .87
Scenario b controllability .86
Scenario c controllability .83
Scenario d controllability .82

Parent responsibility

(eigenvalue=2.5)
Scenario b responsibility .86
Scenario c responsibility .84
Scenario e responsibility .78
Scenario a responsibility .70

 

Table 19 Mother WAQ Attributions about parent role in child behavior factors for
Oppositional/Defiant child behaviors
Factors Individual Variable Labels Factor 1 Factor 2

 

Parental ability to control child behavior
(Eigenvalue=2.7)

Scenario a controllability .87
Scenario b controllability .86 -.15
Scenario c controllability .83
Scenario (1 controllability .82

Parent responsibility

(Eigvenvaluefl. 1)
Scenario b responsibility .86
Scenario c responsibility .25 .84
Scenario e responsibility -.20 .78
Scenario a responsibility .70

 

Reliabilities were good for all mothers’ attributions pertaining to her own role in
child’s behavior, except for attribution about control of child Oppositional/Disobedient
behavior (alpha=.67; see Table 20 for all alpha coefficients).

Table 20 Reliabilities for mothers’ WAQ attributions pertaining to her role in child’s
behavior, by behavior type.

 

 

 

 

Child behavior type Maternal responsibility Maternal control
Inattentive-Overactive .80 .87
Oppositional/Disobedient .83 .67
Prosocial/ Positive .93 NA

 

 

 

 

 

83

Thus, these five additional variables were deemed adequate for exploratory
analyses, except for attribution about control of child Oppositional/Disobedient behavior
(alpha=.67).

RII Attribution Items Pertaining to Parent Responsibility and Control

The parent responsibility and control items were also added to the Recalled

Incident Interview.

Table 21 Maternal RII Attributions about parent role in child behavior factors for
Inattentive-Overactive Behaviors

Factors Individual Variable Labels Factor 1 Factor 2

 

Parental ability to control child behavior
(eigenvalue=l .6)
Behavior 1 controllability .90
Behavior 2 controllability .87

Parent responsrbility

(eigenvalue=l .4)
Behavior 2 responsibility .96
Behavior 1 responsibility .76

 

Table 22 Maternal RII Attributions about parent role in child behavior factors for
Oppositional/Defiant child behaviors
Factors Individual Variable Labels Factor 1 Factor 2

 

Parental ability to control child behavior
(eigenvalue=l .9)
Behavior 2 controllability .88
Behavior 1 controllability .88

Parent responsrbility

(eigvenvalue=l .7)
Behavior 2 responsibility .86
Behavior 1 responsibility .85

 

For maternal attributions about control and responsibility for child
positive/prosocial behaviors, only one factor was found.

Reliability of the R11 measures of maternal control and responsibility were quite
low (alpha’s ranging from .48 to .72; See Table 23). Due to the unsatisfactory
reliabilities obtained on measures of maternal control and responsibility factors using the

R11, this measure was not used to test exploratory analysis 2.

84

Table 23 Reliabilities for mothers’ RII attributions pertaining to her role in child’s

 

 

 

 

behavior, by behavior type.
Child behavior type Maternal responsibility Maternal control
Inattentive-Overactive .48 .72
Oppositional/Disobedient .63 .72
Prosocial/ Positive .65 .64

 

 

 

 

 

Hoza Interactions Questionnaire

Three subscales were used for Exploratory Analysis 2: (1) parent efl'ort to
control child’s behavior, (2) parent general quality as parent (“good parent”) and (3)
parent ability to obtain child behavioral compliance. A score was obtained for each of
these subscales for child compliance and for child non-compliance. The Hoza consisted
of six scenarios (3 compliance and 3 non-compliance); thus, each subscale would be a
composite of 3 items.

An initial factor analysis was conducted with all three potential subscale items per
each of the behavioral scenarios (6 total, 3 of each type); thus, the initial factor analysis
consisted of 18 items. This initial factor analysis resulted in a four-factor structure,
separated by compliance and non-compliance for the first factor, which was a non-
compliance factor composed of “good general parenting” and “parent ability to control
child’s behavior.” The second factor was composed of “parent ability to control child’s
behavior” for compliance items only. The third factor combined items from compliance
and non-compliance scenarios but all items were related to “parent making special effort
to control child’s behavior.” The last factor combined compliance and non-compliance
items related to “general good parenting.” The factor structure is shown in Table 24.
[Footnotez for most measures, initial factor analysis did not yield a useable factor

solution. Factor analyses were then conducted separately by behavior type. The initial

85

factor analysis of the Hoza interaction questionnaire yielded clear and useable results.
Therefore, those factors were retained rather than factors separated by behavior type.
Notably, for comparison to Hoza articles it may be useful to separate by behavior type;
however, the analyses of interest differ from Hoza’s studies and direct comparison would

not be possible anyway].

86

Table 24 Factor structure for Hoza maternal attributions about own parenting and effort

 

 

Factors Variable Labels Factor 1 Factor 2 Factor 3 Factor 4
Non-compliance good parenting

and ability to control child behavior
(eigvenvalue=6.0)

Sen 4 non-comp good .89
Sen 4 non-comp good .86
Sen 4 non-comp ability .85
Sen 5 non-comp ability .84
Sen 2 non-corrrp good .72

Compliance parent ability to control child

Behavior
(eigenvalue=3.0)
Scenario 3 comp good parenting .89
Scenario 1 comp good parenting .80

Scenario 6 compliance good parenting .80

Special effort to control child’s behavior

Non-compliance and compliance

(eigvenvalue=1.7)
Scenario 6 compliance special effort .78
Scenario 5 non-compliance special effort .73
Scenario 3 compliance special effort .72
Scenario 1 compliance special effort .64
Scenario 2 non-compliance special effort .61 .38
Scenario 2 non-compliance special efi‘ort -.33 .55

Good general parenting

(eigvenvalue= 1.2)
Scenario 3 compliance good parenting .32 -.73
Scenario 6 compliance good parenting -.72
Scenario 2 non-compliance good parenting -.71
Scenario 1 corrrpliance good parenting .41 -.48

 

Reliabilities of Hoza subscales were satisfactory (see Table 25). The reliability of
the first two subscales: non-compliance good parenting and compliance ability to control

child had good reliabilities (alpha=.89 and .82, respectively).

87

Table 25 Hoza four factor subscale reliabilities and inter-correlations.

 

 

 

 

 

 

 

 

 

 

 

Non compliance Compliance ability Compliance and General good
good parenting to control child non parenting (comp
Special effort and non)
Non corrrpliance .89
M parenting
Compliance ability -.Ol .82
to control child
Compliance and non .27" .21 .73
Special effort
General good .13 53"“ .28" .78
parenting (comp and
non)
Child Attribution Measures.

Child attributions were measured by an adapted version of Johnston’s Written

Analogue Questionnaire and an adapted version of Hoza’s Interactions Questionnaire.

During initial testing with these measures, it became evident that children under the age

of 10 were having difficulty with the Written Analogue Questionnaire. Although the

measure had been adapted, children appeared to have difficulty. Testers noted that

children did not appear to understand some of the questions and did not appear to be

engaged. It was uncertain whether their lack of understanding was due to lack of

engagement or vice versa.

With three children (ages 7, 9, and 10), the researcher and an assistant worked on

simplifying the structure of the measure, using a forced-choice followed by multiple

choice format. Prior adaptation had been to ask the child to imagine him/herself as the

child in the story and inserting the use of first person rather than third person (you versus

he/she). An additional adaptation was to occasionally ask the child details about the story

which might pertain personally to him or her. For example, for an item that the child was

looking for a piece of lost sporting equipment. Our participants were first asked what

88

 

sports if any they played and that sport with that piece of equipment was inserted into the
item. These changes appeared to increase the child’s engagement in the task. A final
change was to have pictures drawn by an artist to correspond with each item (only two
items did not have corresponding pictures drawn due to an oversight of the author in
communication with the artist). While we were able to collect the child measure from
most children, due to time constraints not all measures were obtained from every child.
Despite this, the sample size was quite good for the child attribution measures. One
hundred and twenty children completed the Written Analogue Questionnaire (W AQ) and
108 children completed the child Hoza. Although the Hoza is a secondary measure (not
primary for hypotheses), it is used as a way to provide some construct validity for the
child WAQ that was created fiom the adult WAQ for this study and has not been used in

prior studies.

Child Written Analogue Questionnaire

As explained earlier, the child analogue questionnaire was adapted from the adult
measure and then adapted into a format which could be better used by younger (ages 6-9)
as well as the older (ages 10-14) in the current study. The data thus yield continuous
ratings from the children with the ability to use dichotomous ratings if the continuous
data did not form useable scales. Analyses revealed a usable continuous variable similar
to the adult version of the WAQ.

As with the adult attribution measures, when all items were included in a factor
analysis, no useable factors emerged. Following procedures with parent measures, factor

analyses were conducted by behavior type. Within the child WAQ, the first four

89

scenarios (A, B, C, D) were clearly about Inattentive-Overactive behaviors. Although
intended as an Oppositional/Disobedient scenario, scenario E (not ending video game and
coming to dinner when told) potentially confounded inattentive/distractible behaviors
with oppositional. Scenario E was run in factor analysis with 10 scenarios (A-D) and the
Oppositional scenario; it did not change results for IO factors and clustered meaningfully
with other OPP items so was kept with OPP ratings. Children rated behaviors on the four
attribution domains. The intemal/extemal domain was worded as “about you” or “not
about you/about the situation.” The controllability domain used the wording “you could
have stopped/done differently” versus “you could not have stopped/done differently.”
The stability domain was worded: you “would do every time” (stable) versus you “would
do differently all the time” (unstable). The globality domain was worded: “only at home”

versus “home and everywhere.”

9O

Child Ratings of Inattentive-Overactive Scenarios
Factor analysis yielded a six-factor solution with three usable factors.

Table 26 Factor analysis of child WAQ attribution ratings pertaining to Inattentive
Overactive child behaviors

 

 

Factors Labels Factor 1 Factor 2 Factor 3 Factor 4 Factor 5 Factor 6
Globality (eigenvalue=2.8)
Scenario D globality .80 -. 14 -.14 -.15 .11

Scenario B globality .78
Scenario C globality .77 .20

Scenario A globality .66 -.17 .10 .17
Controllability (eigenvalue=2.3)

Scenario D control -.81 -. l4

Scenario A control -. 14 -.79 -. 15 .26

Scenario B control -.68 .11

Scenario C control .21 -.57 -.19 -.15
Locus 1 (eigenvalue=l.7)

Scenario D locus .87 .13 -.27

Scenario A locus -.14 ' .60 -.15 .30 .21

Scenario C locus .43 -.35 .29 -.36
Locus 2 (eigenvalue=l.3)

Scenario B locus .12 -.83

Scenario C stability .13 .20 .65 .37 .25
Stability (eigenvalue=l .1)

Scenario D stability .14 .18 .79 -.18

Scenario B stability .12 -.14 .73 .14

Scenario A stability .12 .91

 

The first two factors were not heavily cross-loaded with other items (only one
item cross-loading onto the first two factors at greater than .25). Although the third
factor was cross-loaded with other items, it was still retained. Unsurprisingly, in the final
factor analysis, the first two factors held whereas the third factor was less clean (as shown
in Table 26). It loaded onto a fourth factor until two of the cross-loading items were

deleted.

91

Thus, a three-factor solution was retained, providing child ratings of globality,
controllability and locus pertaining to Inattentive-Overactive behaviors. The three-factor
solution retained is shown in Table 27.

Table 27 Factor analysis of child WAQ attribution ratings pertaining to Inattentive
Overactive child behaviors

 

 

Factors Individual Variable Labels Factor 1 Factor 2 Factor 3
Globality (eigenvalue=2.4)
Scenario D globality .79
Scenario B globality .77 -.21
Scenario C globality .76 .13 -.13
Scenario A globality .71 .12 .20
Controllability (eigenvalue=l .8)
Scenario D controllability .84
Scenario A controllability -. 15 .75 .15
Scenario C controllability .17 .65 -. 17
Scenario B controllability .11 .64

Locus (eigenvalue=l .2)
Scenario A locus -.10 .83
Scenario D locus .72

 

For child attributional ratings of Inattentive-Overactive behaviors, only globality
and controllability had adequate reliabilities (alpha=.77 and .69, respectively). The

reliability of the locus ratings was unsatisfactory (alpha=.39 for locus).

Child Ratings of Oppositional/Disobedient Scenarios
Only two scenarios pertained to Oppositional/Disobedient behaviors. An initial
factor analysis yielded a four factor solution (see Table 28) with stability items cross
loading with locus and controllability (greater than .31). The stability items were deleted

and a usable three-factor solution emerged (see Table 29).

92

Table 28 Factor analysis of child WAQ attribution ratings pertaining to
Oppositional/Disobedient child behaviors

 

 

Factors Individual Variable Labels Factor 1 Factor 2 Factor 3 Factor 4
Locus (eigenvalue=2.0)
Scenario F locus .77 .14
Scenario E stability .65 -.31 .10
Scenario E locus .55 -. 14 .13
Globality (eigenvalue=1.2)
Scenario E globality -88 .14
Scenario F globality .17 -81 -.14
Controllability (eigenvalue=1 . 1)
Scenario B controllability -.12 .85 .20
Scenario E connollability .42 .19 .57 -.35

Stability (eigenvalue=1 .1)
Scenario F stability .17 .12 .92

 

Table29 Factor analysis of child WAQ attribution ratings pertaining to
Oppositional/Disobedient child behaviors

 

 

Factors Individual Variable Labels Factor 1 Factor 2 Factor 3
Locus (eigenvalue=1 .9)
Scenario F locus .83
Scenario E locus .70 -.13
Globality (eigenvalue=1 .2)
Scenario E globality -88
Scenario F globality .15 -79
Controllability (eigenvalue=1.0)
Scenario E controllability -. l9 -. 17 .89
Scenario F controllability .38 .19 .62

 

Reliabilities were poor for locus and controllability(a1phas=.49 and .36
respectively) and borderline for globality (alpha=.64). Only the globality factor

reliability was adequate for retention as a factor.

Child Ratings of Positive/Prosocial Scenarios
The factor analysis of child attributional ratings of positive/prosocial behaviors
yield three-factors. Globality and stability factors emerged as one factor with stability

items cross loading (.41 with the locus factor). The stability items were deleted and a

93

clean three-factor solution (with no cross-loadings greater than .25) emerged (see Table
30).

Table 30 Factor analysis of child WAQ attribution ratings pertaining to
Positive/Prosocial child behaviors

 

 

Factors Individual Variable Labels Factor 1 Factor 2 Factor 3
Globality (eigenvalue=2.6)

Scenario H globality -.85 -. 19

Scenario G globality -.81 .19 -.13
Controllability (eigenvalue=1.6)

Scenario H controllability .88

Scenario G controllability .87
Locus (eigenvalue=1 .0)

Scenario G locus .83

Scenario H locus .80

 

Reliability for controllability was adequate (alpha=.7l) but borderline for
globality and locus (alphas=.56 and .52, respectively).

As shown in Table 31, child attributional ratings of globality for Inattentive and
Disobedient behaviors are significantly related. Controllability ratings are highly
correlated across all three behavioral types. Globality was correlated for rating pertaining

to 10 and Opp behaviors.

94

Table 31 Inter-correlations among child WAQ attribution ratings, data reliabilities on the

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

diagonal
IO 10 IO OPP/D OPP/D OPP/D POS/Pro POS/Pro POS/Pro
globality locus control globality locus control mbahty locus control
10 .77
_globality
IO locus .07 .39
IO -.21"' .06 .69
control
OPP/D 69*" .01 -.26** .64
_globality
OPP/D -.23"' -.16+ .19“ -.23* .49
locus
OPP/D -.10 -.02 .40**"‘ -.13 .22" .36
control
POS/Pro .09 -.l8+ -.24* .16 .13 .23“ .56
Lglobality
POS/Pro -.30""" -.06 .06 -.l8+ .23‘ .21“ .27" .52
locus
POS/Pro -.20"‘ -.06 .30“ -.18" .l l .40"" .08 .20‘ .71
control
"“p<.001, ”p<.01, ‘p<.05, +p<0.1
Child HOZA

The Hoza measure contains six scenarios, three pertaining to compliant behaviors

and three pertaining to non-compliant behaviors. Children were asked to rate five

possible reasons why they might do the behavior described in each scenario. Ratings

were made on a five-point scale (1=very true to 5=not true at all). The five “reasons”

included: a) child mood, b) child effort or lack of effort, c) child ability or lack of ability

to control self, (1) maternal effort or lack of effort, and e) maternal ability or lack of

ability to “get you (child) to obey her.”

A factor analysis with all 30 items did not converge into a usable pattern matrix.

Following procedures throughout data reduction, factor analyses were conducted by

behavior type (in this case, compliance versus non—compliance).

95

 

Behavioral Compliance
For compliant behavioral scenario, a five factor solution emerged but with heavy
cross-loadings (see Table 32). Maternal effort and control cross-loaded as did child effort
and control.

Table 32 Factor analysis of child Hoza attribution ratings pertaining to Compliant
behaviors

 

 

Factors Labels Factor 1 Factor 2 Factor 3 Factor 4 Factor 5
Child mood and effort (eigenvalue=4. 6)

Scenario 6 child effort .86

Scenario 6 child mood .65 .13 -.30

Scenario 3 child effort .54 -.22 .13 -.21
Mother effort (eigenvalue=1.6)

Scenario 1 mom effort .87 .10

Scenario 3 mom effort .68 -.10 -.19

Scenario 6 mom effort .33 .47 -.41

Mother control (eigenvalue=1 .5)

Scenario 6 mom control -.27 -.82
Scenario 3 mom control -.81
Scenario 1 mom control -.17 .21 -.64
Child control (eigvenvalue=] .2)
Scenario 3 child control .88
Scenario 1 child control -.11 .81
Scenario 6 child control .51 .60 .25
Child mood and effort (eigvenvalue=1.0)
Scenario 1 child mood -.12 -.76
Scenario 3 child mood .14 .16 -.72
Scenario 1 child effort .29 -.13 -.13 .13 -.58

 

*cross loadings greater than .25 suppressed

When items which cross-loaded at greater than .25 were deleted, a three factor

solution emerged but again items were heavily cross-loaded with only 3 items not cross-

loaded greater than .25 (see Table 33).

96

Table 33 Three-factor solution of child Hoza attribution ratings pertaining to Compliant
behaviors

 

 

Factors Individual Variable Labels Factor 1 Factor 2 Factor 3
Child control (eigvenvalue=3.1)
Scenario 1 child control .79
Scenario 3 child control .79 -.16
Scenario 1 mom effort -.29 .80

Unnamed (eigvenvalue=1.4)

Scenario 3 mom effort .68 -.23
Scenario 3 child mood .47 .56
Scenario 1 child mood .33 .55
Mother control and child effort (eigvenvalue=l .2)
Scenario 3 mom control -.80
Scenario 1 mom control -16 -.71
Scenario 3 child effort .28 .10 -.65
Scenario 6 child effort .24 -.51

 

*cross loadings greater than .25 suppressed

When additional cross-loaded items were deleted, a final solution (shown here in
Table 34), consisting of maternal control and child control, emerged.

Table 34 Two-factor solution for child Hoza compliance items

 

 

Factors Individual Variable Labels Factor 1 Factor 2
Mother control (eigvenvalue=2.1)

Scenario 1 mother control .81

Scenario 3 mother control .81

Scenario 3 mom effort .56

Child control (eigvenvalue=1.2)
Scenario 1 child control -92
Scenario 3 child control -.87

 

*note: no items cross-loaded at greater than .25.

Reliability was inadequate for child’s attributions about maternal control
(alpha=.55) but adequate for rating of own control (alpha for child control=.75). If the
factor maternal control was created without the maternal effort item included, reliability
increased significantly (alpha for mother control composed of two control items=.6l).

Child rating of self-control and child rating of maternal control of child’s compliance

97

behavior were significantly correlated although the size of the correlation was relatively

small (F22, p<.05).

Behavioral Non-Compliance
When analyzing items pertaining to child non-compliance, no factor solution
emerged in less than 25 rotations. Four factors had eigenvalues greater than 1 but no
pattern matrix emerged even when tried to force 4 factor solution. Thus, no factors were

created for the child Hoza non-compliance items.

Summary Regarding Child Attribution Measures

The child WAQ factor analysis yielded usable factors with satisfactory
reliabilities. Factors retained for analysis were: globality and controllability ratings for
Inattentive-Overactive behaviors; globality rating for Disobedient behaviors, and
controllability for Positive behaviors. The child Hoza measure only yielded one factor
with satisfactory reliability: child rating of own ability to control behavior in relation to
scenarios about behavioral compliance. Child’s rating of mothers ability to control
child’s compliance had borderline but still usable reliability (alpha=.61). It was not
possible to obtain a factor structure for child ratings of non-compliance behaviors on the
Hoza measure.

The child WAQ subscales are compared to the child Hoza through an examination
of correlations (see Table 35). The two measures were not highly related; however, only
a few subscales were found reliable. Even those subscales with acceptable reliability

were not significantly related to child ratings of self-control. Despite the lack of relation

98

between the Hoza and WAQ scales, the intercorrelations of items within scales suggest
consistency within measure. However, lack of consistency across the measures suggests
the need for additional psychometric and validity tests for both measures. Unexpectedly,
child rating of self-control on Hoza compliance scenario was negatively related to child
rating of internal locus on WAQ positive behavior scenario.

Table 35 Correlations between child WAQ and Hoza ratings

 

 

 

 

 

 

 

 

 

 

 

 

 

WAQ subscales Hoza child ratings of Hoza child ratings of
Child self-control of Mother control of
compliance behaviors corrrpliance behaviors
IO glo_bality .17+ -.13
IO locus .01 .06
IO control .01 -.13
OPP/Djlobality .12 -.ll
OPP/D locus -.14 .15
OPP/D control -.11 .05
POS/Pro -.l 1 -. 14
Jammy
POS/Pro locus -.22* -.1 l
POS/Pro -.16 -.04
control

 

mp<.oor, “p<.01, “p<.05, +p<0.1

Child WAQ controllability rating of IO behavior and child Hoza rating of

controllability were deemed the two most reliable and useful factors for test of

hypotheses. Child WAQ ratings of IO globality, OPP/Disobedient globality and control,

and POS control subscales were retained as well due to their satisfactory reliabilities

and/or use for comparison to Hoza measures. Despite the lack of correlations between

the two measures, correlations were good within each attribution measure.

Note regarding the Creation of Covariate Variables

No data reduction was conducted for creation of the co-variable variables. As

noted in method section, reading disability was created through comparison of WISC and

WIAT scores. Diagnosis of Oppositional Defiant Disorder was based on the DISC

interview.

99

 

Primary Hypotheses
Hypothesis 1 .' Maternal attributions and child ADHD

Hypothesis 1a. Maternal attributions about child behavior will differ based on
ma; child diagnosis (child diagnostic group). Because a diagnosis of ADHD given to a
child was theorized to affect parental attitudes about one’s child’s behavior, mothers’
attributions about child behaviors were examined in terms of group comparisons.

Prediction la. ADHD diagnosis assigned to child was expected to be associated
with parent stable, global, uncontrollable, and internal attributions. That is, parents of
children with ADHD were expected to endorse more stable-global, uncontrollable, and
external attributions compared to attributions made by parents of children without
disorders. A difference was expected on all four dimensions.

Hypothesis 1b. Maternal attributions about child behavior will differ based on

mothers’ beliefs as to whether or not her child has ADHD.

 

Prediction lb. Parents’ belief that the child’s behavior warranted an ADHD
diagnosis or that the child has undiagnosed ADHD was also expected to be associated
with parent stable, global, uncontrollable, and extenral attributions about child disruptive
behaviors. That is, parents who believed that their child has a diagnosis of ADHD or
believed that their child had undiagnosed ADI-II) (whether or not our through our testing
we discovered the child met criteria) were expected to endorse more stable, global,
uncontrollable, and external attributions about their children’s misbehaviors compared to
attributions made by parents who believed that their children did not have ADHD

(whether or not that turned out to be true). Because there may or may not be

100

isomorphism of parental belief about child diagnosis and actual diagnosis, this was tested

as a separate analysis.

Hypothesis 2: Maternal attributions and role adjustment

Hypothesis 2. Maternal attributions were expected to be related to maternal role-
adjustment.

Maternal cognitions and beliefs about family problems have been related to
maternal adjustment to those problems and to maternal role-specific adjustment. It was
expected that this relation would depend upon child diagnosis. However, two competing
hypotheses were tested for thoroughness. That is, it was hypothesized that in
dimensional analyses, matemal attributions about their child’s behavior would be related
to their own role-specific adjustment, with stable, global, controllgble. and internal
attributions being related to greater role stress. This prediction that attributions would be
related to controllable attributions was based on general literature (Dix & Grusec, 1985)
with parents of non-disordered children. Such a finding would indicate that regardless of
child diagnosis, controllable attributions for nrisbehaviors lend toward parental stress.
The current author expected that the dimensional relation between attributions and
adjustment would be mediated by child symptoms and moderated by child diagnosis;
thus, in addition to testing for the simple correlation, this hypothesis was explored
controlling for controlling for child behaviors and then child diagnosis. If the relation
between maternal attributions and adj ustment were mediated by child behaviors, the
importance of diagnosis would indirectly be supported. If maternal attributions were

found to be independently related to role stress (with child behaviors controlled), then the

101

attributional profile would appear to not depend upon diagnosis. Notably, the moderation
hypothesis would best test the role of diagnosis in the relation between maternal
attributions and adjustment.

Prediction 2a. Attributions of intemality, stability, globality, and controllability
about child misbehavior were expected to relate to high levels of parental stress, when all
subjects were included in dimensional analyses. 2b) This relation was expected to
mediated by child behavioral severity. 2c) Alternatively, this relation could have been
independent of child behavior, which would support an independent relation between
maternal attributions about child behavior and maternal stress.

Note re: analyses: For hypothesis 2b, parent perception of child severity was
proposed for this analysis. There are multiple considerations that led to this decision. 1)
Parents may be stressed by the fact that they think their child behaves badly or because
they think the behavior has a particular cause. Whether parents’ attributions determine
stress could thus be checked by covarying parent perception of the child behavior. 2) To
test whether parent stress was due to actual child behavior or because of their attributions
would ideally be tested with observational data that was beyond the scope of the

proposed study.

Hypothesis 3:Moderation by child diagnosis
Hypothesis 3) In further understanding the relation between maternal attributions
and role adjustment, this relation was expected to depend upon (be moderated by) child
diagnosis. [The interaction between parent attributions and child diagnosis was expected

to be significant in predicting parental adjustment].

102

Exploratory Analyses.
Exploratory Analysis I : Maternal Attributions and child dimensional behaviors

Exploratory Analysis 1. It was proposed to examine maternal attributions in
relation to severity of child behavior (collapsed across diagnostic groups); however, this
analysis was part of the analyses covered in Hypothesis 2b.

Because no known current studies have examined parent attributions about child
misbehaviors fi'om a dimensional perspective and parent attributions may vary not just by
child diagnosis (the primary question of interest in the proposed study) but also by
severity of child behaviors, it is reasonable to test this.

Prediction Exp Analysis 1. In order to understand how mother’s beliefs relate to
child disruptive behaviors (inattention, hyperactivity, and aggression), examination of
significant relations between maternal attributions and dimensional ratings of child
behaviors were conducted. To compare mother versus teacher ratings (and versus other
parent ratings of child behavior) would allow a possible separation of mother’s
perception of child behavior and actual child behavior. It should be noted, however, that
the literature shows that differences do exist across raters when examining child ADHD
behavior. No conclusions could be drawn fiom these analyses. Maternal attributional
ratings of intemality, stability, globality, and uncontrollability would be expected to be
related to greater severity of child hyperactivity, inattention, and aggression. No

predictions were made about ratings by mother versus teachers (or versus fathers).

103

Exploratory Analysis 2: Maternal responsibility and control
In prior studies, attributions about child controllability have been found as key to
parental adjustment (Dix & Grusec, 1985), the extent to which parents feel responsible
for their child’s behavior and the extent to which they feel both responsible and able (or
unable) to control their child’s behavior might be important in their adjustment. An
additional exploratory analysis was that mothers’ experience of responsibility would
interact with her perceived ability to control her child’s behavior, and that this interaction

would predict parental role adjustment.

Exploratory Analysis 3: Child attributions

Child attributions about their own disruptive behaviors were expected to differ by
diagnostic status and be correlated with severity of own behaviors. The bulk of the
attribution and ADHD has examined attribution in relation to social or academic task
outcomes, few studies (none known) have examined children’s attributions in relation to
their disruptive behaviors. There is a large body of literature on attributions and
aggression, mostly social attributions (that is, attributions about others rather than about
one’s own behavior). To date, it is unclear whether children with ADHD tend to make
more internal attributions (Carlson, Mann, & Alexander, 2000; studied effects of reward
and response cost in relation to attributions and performance) or external attributions for
negative events (Milich, 1994; studied attributions pertaining to failure on academic
tasks). Very little if anything is known about such attributions made by children with
ADHD in relation to disruptive behaviors and specifically to behaviors which they

(versus another actor) might engage in. In order to shed light on the types of attributions

104

made by children about disruptive behaviors, the current study attempted to measure
multiple attributional ratings about Inattentive-Overactive, DisobedienUnon-compliant,
and Positive/prosocial behaviors.

Exploratory Prediction 3a. Child attributions would differ based on diagnostic
grouping. Specifically, children with ADHD were expected to ascribe disruptive
behavior to more external, stable, global, and uncontrollable causes compared to the
attributions made by non-disordered peers. The primary analyses were between ADHD
and non-disordered children. Follow-up analyses examined possible differences between
children with ADHD-inattentive type (ADD), ADHD-combined type (ADHD-C) and
children with subthreshold problems.

Exploratory Prediction 3b. Child attributions would be related to their adult-rated
behaviors. Child attributional ratings of extemality, stability, globality, and
uncontrollability were expected to correlate with greater severity of adult-rated child

hyperactivity, inattention, and aggression (collapsed across child diagnostic groups).

105

CHAPTER 4
RESULTS FOR STUDY 1

Maternal Attributions
Hypothesis 1 .' Maternal attributions and child disruptive behaviors
Hypothesis 1 a: Maternal Attributions and actual child diagnosis

For between group comparisons, child diagnostic group served as the independent
variable with each attribution domain serving as dependent variables. One-way analysis
of variance was computed with the independent variables having two levels (ADHD and
controls). The ADHD group was composed of the ADHD-Combined and ADI-ID-
Inattentive cases.

As shown in Table 36, mothers of children with ADHD made attributions that
were more stable, global, and uncontrollable compared to mothers of children without a
diagnosis of ADHD. For example, for the Inattentive-Overactive type of behavior,
mothers of children with ADHD made attributions which were significantly more stable
(F[l, 81]=12.6, p<.001), global (F[1.81]=18.1, p<.001), and uncontrollable (F[1,81]=6.1,
p<.05)7.

This finding was true for each of the three behavioral types examined in the
Written Analogue Questionnaire; that is, for Inattentive-Overactive,
Disobedient/Oppositional, MPositive/Prosocial behavior types, mothers of children
who had diagnosable ADHD made more stable, global, and uncontrollable attributions
compared to mothers of children without such a diagnosis (see Table 36). No significant

difference was found on the attribution domain of locus (internal versus external).

 

7 Note: Out of the possible n=91(50 for ADHD groups combined and 41 for control group), only 81
mothers completed the WAQ dropping the sample size to 81 for these analyzes).

106

Table 36 Test of differences in attributions made by mothers of children with ADHD
diagnosis versus mothers of children without a behavioral disorder, using the WAQ
measure that describes child behavior problems.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Control ADHD F statistic P value Value on attrib Attributions about
Mean any type (df) measure ADHD group in
(sd) Mean comparison to
(89) control
WAQ 3.0 4.3 F(1,8l)=l2.6 P<.001 1=unstable ADI-ID more
IO (1.6) (1.6) 10=stable stable
stability
WAQ 4.0 5.8 F(1,81)=18.1 P<.001 l=specific ADHD more
IO (1.9) (1.9) 10=global global
__globality
WAQ 5.8 6.2 F(1,81)=.55 Ns 1=extemal No sig difference
IO locus (2.3) (1.9) 10=internal
WAQ 8.1 6.9 F(1,81)=6.1 P<.01 1=uncontrollable ADHD less
IO (1.8) (2.4) 10=controllable controllable
control
WAQ 2.7 4.2 F(1,8l)=13.2 P<.001 l=unstable ADI-ID nrore
OPP (1.6) (2.1) 10=stable stable
stability
WAQ Not Not 1=specific Not able to test
OPP available available 10=global
_globality
WAQ 6.8 6.9 Fl,8l)=.05 Ns 1=extema1 No sig difference
OPP (2.0) (2.0) 10=internal
locus
WAQ 8.6 7.6 F(1,8l)=6.5 P<.01 1=uncontrollable ADHD less
opp (1.6) (2-0) 10=controllable controllable
control
WAQ 2.7 4.2 F(1,8l)=13.2 P<.001 l=unstable ADI-ID more
Positive (1.6) (2.1) 10=stable stable
stability
WAQ 3.5 5.5 F( l,81)=18.9 P<.001 1=specific ADHD more
Positive (1.9) (2.1) 10=global global
_globality
WAQ 6.8 6.9 F(1,8l)=.05 Ns l=external No sig difference
Positive (2.0) (2.0) 10=internal
locus
WAQ 8.6 7.6 F(1,8l)=6.5 P<.01 l=uncontrollable ADHD less
Positive (1.6) (2.0) 10=controllable controllable!
More
control uncontrollable

 

I"IO=Inattentive-Overactive type of behavior

*OPP=oppositional/disobedient type of behavior

107

 

Controlling for child aggressive behavior

When controlling for child aggressive behavior, the only group differences for
maternal attributions about children’s ADI-ID behaviors was on the attribution of locus.
When exarrrining child prosocial behaviors, a group difference was found pertaining to
maternal attributions of control (again even with child aggressive behaviors controlled).

Controlling for medication status on subset of sample

Although not originally proposed, for the limited number of subjects for whom
medication status was available (n=64), group differences on maternal attributions were
analyzed controlling for medication status (child regularly taking medications for ADHD
or not). Using, ANCOVA analyses, no significant affects were found when controlling
for medication status.

Subtype Analyses

A unique contribution of the current study was to examined findings by subtype
of ADHD. Thus, after the initial analyses, a one-way analysis of variance was computed
comparing with the independent variable having four levels (ADHD-C, ADD,
subthreshold, and controls). As shown in Table 37, most often differences were between
the ADHD-C and control group; however, group differences were found between each
type of “disordered” group (ADHD-C, ADD, subthreshold) and the control group for
maternal attributions about 10 behavioral globality. Also, maternal attributions differed
by type of ADI-ID diagnosis (ADHD-C vs. ADD) for WAQ stability of Oppositional
behavior, WAQ stability and globality of positive behavior. Maternal attributions

differed between mothers of children with ADD versus mothers of children with

108

subthreshold ADHD (either type) symptomatology for WAQ attributions related to locus
of positive behavior and locus of Oppositional behavior.

These analyses revealed that group differences were frequently but not
exclusively driven by differences between ADHD-Combined subtype and controls. A
notable finding was that in the four-group comparison, differences were foungfirr
mothers’ attribution of locus pertaining to Oppositional and Positive behaviors (but not
pertaining to Inattentive-Overactive behaviors). For this group, children with ADD-
inattentive type differed from children with subthreshold ADHD symptomatology.
Mothers’ ratings of the behavior of their children with ADD-inattentive were more
internal compared to the ratings made by mothers of children with subthreshold
symptoms.

When controlling for child aggression, only maternal globality attributions about

child prosocial behaviors remained significantly different across subtypes.

109

Table 37 Results of maternal attributions about child behavior, testing for group
differences when comparing the independent variable of child diagnosis with four levels
(ADHD-C, ADD, subthreshold, and controls), dependent variable=attribution domain by

 

 

 

 

 

 

 

 

 

 

 

 

 

 

behavior type
a b c
Control ADHD- ADD - Sub- F stat p-value Value on Conclusions
Combined Inattartive threshold ((11) attrib
ADHD measure
WAQ IO 10' 4.6 3.5 3.8 5.5 P<.001 l=unstable Only control and
stability (1.6) (1.7) (1.1) (1.6) (3,115) 10=stable ADHD-C differ
WAQ to 4.0 lbc 59' 555 5.3 c 6.4 P<.001 laspecific Each
819119. Includlnzkl
differ from
control”
WAQ [O 5.8 6.1 6.3 5.7 0.3 n.s l=extanal No group
locus (2.3) (1.9) (1.9) (2.1) (3,115) 10=inter'nal differences were
found
WAQ 10 8.1 ' 6.6 ‘ 8.0 7.1 3.6 P<.01 l-uncontro Only control and
' ' 10=controll
able
WAQ 2,73 4.7“” 2,0" 3.8 8.5 P<.001 l=unstable ADHD-C and
OPP (1.5) (3,1 14) 10=stable control differ
. . .4
stability (16) (21) (1 ) ADHD-Cand
ADD difl‘a'
WAQ 6.8 6.5 82 ° 6 _2 ° 3.0 P<.05 l=external ADD and
OPP (2.0) (1.9) (2 1) (2 1) (3,114) 10=intcrrral subthreshold
locus ' ' differ
WAQ 8.6 ‘ 7.3 ' 8.4 7.7 3.8 P<.01 l=uncontro ADHD-C and
control ' ' 10=controll
able
WAQ 2;, . 4;, sh 2.9 b 3.8 3.0 P<.05 1=unstable ADHD-C and
Positive (1 .5) (3,1 14) 10=stable control differ
. . .4
stability (1 6) (2 I) (l ) ADHD-C and
ADD differ
WAQ 35 a 5.6 ab 53 b 4.5 3.8 P<.0l l=specific ADHD-C and
Positive (1.8) (3,] l4) 10=global control differ
.9 2. 2.6
globality (1 ) ( 0) ( ) ADHD-C and
ADD differ
WAQ 6.8 6.5 8.1 e 6.2 e 8.7 P<.001 l=euternal ADD and
Positive (2.0) (1.9) (2 l) (2 l) (3,1 14) 10=intanal subthreshold
locus ' ' differ
WAQ 8.6' 73 9 8.4 7 .7 6.8 P<.001 l=uncontro ADHD-C and
control ' ' 10=controll
able

 

 

 

 

 

 

 

 

 

 

l’IO-Inatter'rtive-Overactive type of behavior

‘OPP=oppositional/disobedient type of behavior
it indicates significant difference between control and adhd combined groups
b indicates significant difference between control and add inattentive groups
c indicates significant difference between control and “subthreshold” groups
(1 indicates significant difference between adhd and add group

 

e indicates significant differences between add and subthreshold groups
f indicate significant difference between adhd and “subthreshold groups

110

In summary, mothers of children with ADHD viewed their children’s behavior as
more stable, global, and uncontrollable than did mothers of children without a diagnosis
of ADHD. These results were largely due to attributions by mothers of children with
ADHD-Combined type. Results did not remain significant when controlling for child
aggression.

Results were largely similar when using the Recalled Incident Interview (see

Appendix C), although sub-group differences (see Appendix D) were not as consistent.

Hypothesis 1b: Maternal Attributions and beliefs about diagnosis

Maternal attributions were examined in relation to beliefs about a possible ADI-ID
diagnosis. Out of 126 subjects, 109 mothers answered a question as to belief about
child’s ADHD diagnosis. Each mother was asked: “do you believe that your child has
ADHD, whether or not you have been told he or she has ADHD?” Pom-seven mothers
answered, “yes”, they believed their child to have ADHD (whether or not he/she had
been diagnosed). Sixty-two mothers answered that no, they did not believe their child to
have ADHD (whether or not he/she had been diagnosed). Only one mother indicated that
her child had been diagnosed with ADHD but that she did not agree with the diagnosis.
When looking at diagnostic classification, only 75 (of those 109) children were classified
as ADHD or normal control by our study procedure. (The others were subthreshold). Of
those 75 mothers (whose child was classified as ADHD or non-ADHD) and who
answered the item regarding her belief about her child’s diagnostic status, 43 had children

whom we classified as meeting criteria for ADHD and 32 had children whom we

111

classified as non-ADHD subjects. Twenty-nine had children with subthreshold ADHD

 

 

 

 

 

 

 

 

 

 

symptoms.
Table 38 Cross-tabulation of child dia osis with mothers’ belief about diagnosis
Actual child Believe Believe Believe Believe Totals
diagnosis based child has child has child does child does
on strrdy testing ADI-ID ADHD and not have not have
child has ADHD and ADHD
prior no ADHD although
ADHD dx dx in past was given
that dx
Control 1 O 33 0 32 32
ADHD-C 7 l 8 9 0 32 43
ADD-inattn 3 2 5 O 10
ADHD-hyp o 1 o o r
Subthreshold 8 7 14 1 29 29
Totals 19 28 61 l 109
47 62 109

 

 

 

 

 

 

 

When using this “belief about ADHD” variable to separate mothers into two
groups (believe child has ADHD vs. believe child does not have ADHD) results were
qualitatively similar to those when using actual child diagnosis (see Tables 36 and 37).
That is, using the “belief” item, the pattern of results pertaining to maternal attributions
when looking at differences based on belief about ADHD diagnosis (believe ADHD vs.
believe no ADHD) (see Table 39) were similar to the pattern of results when examining
group differences based on actual diagnosis (Table 36). For example, mothers who
believed their child had ADHD answered qualitatively similarly on attributions related to
Inattentive-Overactive behaviors and Disobedient behaviors when compared to mothers
of children who actually had ADHD. Their attributions were more global, stable, and
uncontrollable than mothers who did not believe their child had ADHD.

Qualitatively, two findings were different from results found when using actual
child diagnosis. Mothers who believed their child to have ADHD ascribed positive

behaviors as (a) more under the child’s control (more controllable) and (b) more specific

112

to a particular situation, than mothers who believed their child to be free of ADI-ID.
Recall that for actual diagnosis, mothers of children with ADHD ascribed positive
behaviors as (a) less under the child’s control and (b) less specific (more global) to a

particular situation.

113

Table 39 Test of differences in attributions made by mothers of children who indicated
that they “believe their child has ADI-ID” (whether or not our testing confirmed or
disconfirmed this) versus mothers who believe their child “does not have ADHD);
attributions as measured by the WAQ standardized scenarios in which to imagine own

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

child enggn’ g
Mother Mother F statistic P value Value on attrib Attributions
answered that answered (df=l , 105) measure when believe
she believe that she child has
child has believe ADHD
ADI-ID Child does consisted to
Mean not have belief child
(sd) ADHD does not
Mean
(sd)
WAQ 10 4.3 3.3 7.8 P<.01 l=unstable More stable
stability (1.7) (1.7) 10==stable
WAQ 10 5.9 4.6 11.5 P<.001 l=specific More global
obality (2.1) (1.9) 10=global
WAQ 10 5.8 6.0 .09 n.s. l=external No sig
locus (1.9) (2.2) llhinternal difference
WAQ 10 6.7 7.8 7.0 P<.01 1=uncontrollabl More
control (2.4) (2.0) e uncontrollable
10=controllable
WAQ ODD 4.4 3.1 12.5 P<.001 l-unstable More stable
stability (1.8) (1.8) 10=stable
WAQ ODD Not available Not available Not Not l=specific NA
globality available available 10=global
WAQ ODD 6.4 6.7 0.4 n.s. l=external No sig diff
locus (2.0) (2.2) 10==internal
WAQ ODD 7.3 8.4 9.7 P<.01 l-uncontrollabl More
control (2.0) (2.7) e uncontrollable
lO-eontrollable
More WAQ 4.4 3.1 12.5 P<.001 l=unstable More stable
Positive (1.8) (1.8) 10=stable
stability
WAQ 5.3 4.2 8.0 P<.01 l=specific More
Positive (2.0) (2.1) lO-global specific/Less
_globality ma!
WAQ 6.4 6.7 0.42 n.s. l=external No sig diff
Positive (2.0) (2.2) lO-internal
locus
WAQ 7.3 8.4 9.7 P<.01 l=uncontrollabl More
Positive (2.0) (l .7) e controllable
control 10=controllable
Overall, hypothesis 1 was supported.

114

 

Additional analysis: maternal attributions and dimensional child behaviors

As shown in Table 40, maternal attributions were related to child behaviors such

that greater endorsement of stability and globality attributions was related to higher levels

of child disruptive behaviors as rated by both mothers and teachers. Maternal attributions

of control were negatively correlated with child disruptive behaviors, again using both

mother and teacher ratings of child behaviors.

Table 40: Correlations between maternal attributions and child behaviors

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Morn Morn Morn Teacher Teacher Teacher
rating of rating of rating of rating of rating of rating of
child child child child child Hyp child
Inattn Hyp ODD Inattn ODD
WAQ 10 .41*** .34*** .35*** .27" .28" .17+
stability
WAQIO .47*** .43*** .44*** 34*" 25* .23+
flbality
-WAQ 10 .08 .08 .15+ .09 .09 -.01
locus
WAQ 10 -.29** -.20* -. l O -.20* -.23* .01
control
WAQ ODD .43*** .44*** .46*** .25* .32" .30"
stability '
.WAQ ODD Not available Not available Not available Not available Not available Not available
_globality
IWAQ ODD -.O4 .01 .05 .07 .07 .03
ocus
WAQ ODD -.31*** -. 15 .00 -.20* -. 18+ -.07
control
M0" WAQ .43*** .44*** .46*** .25* .32" .30"
Positive
stabilig
WAQ .27*** .49**"' .51*** .32" 34*" .34"
Positive
_globality
WAQ -.04 .01 .05 .07 .07 .03
Posrtrve locus
WAQ -.31*** -.15 .01 -.20+ -.18+ -.07
Positive
control

 

115

 

Hypotheses 2 and 3. Maternal Attributions and Role Adjustment

Hypothesis 2a was that attributions of intemalig, stability, globalig, and
controllability about child misbeh_avior would correlafi with maternal stress (or low
levels of satisfaction) when all subjects were included. In essence, this hypothesis aimed
to replicate early findings in parents of non-disordered children (Grusec & Dix, 1985).
Hypothesis 2b was that this relation would be independent from child behavior problems.
Hypothesis 20 was that this relation would be mediated by child behavior. Hypothesis 3
was that the relation between parent attributions and role adjustment would be moderated
by child behavior problems or diagnosis. Notably, Hypothesis 2b and 2c are alternative
hypotheses to be explored for each of the disruptive behavior types
(inattentive/overactive and oppositional/disobedient). It was uncertain what findings
would reveal regarding these contrasting hypotheses and for each behavior type. Based
on the literature, Hypothesis 3 was expected to be the clearest result. Thus, the
expectation was that the relation would be moderated by child diagnosis but as results are

exploratory all analyses are included.

116

ijothesis 2a: Dimensional relation between attributions and stress.

As hypothesized, greater stability and globality were associated with lower levels

of role satisfaction and higher levels of role stress (see Table 41).

Table 41 Correlations between maternal attributions (using scenario measure, WAQ) and
arent role stress and satisfaction (2-tai1ed)

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Mothers Perception of Mothers PSI Total Stress

Parenting Performance
WAQ IO stability -.23* 34*”
WAQ IO globality -.22* 43*"
WAQ IO locus -.09 .15
WAQ 10 control .03 -.15
WAQ ODD stability -.34"‘" 45*"
WAQ ODD globality NA NA
WAQ ODD locus -.12 .01
WAQ ODD control .09 .09
WAQ Positive stability -.34"'*"' 45*“
WAQ Positive globality -.31“"“* .47"'*"'
WAQ Positive locus -.12 .01
WAQ Positive control .09 .16

 

”‘p<.001, ”p<.01, “p<.05, +p<.1

Table 42 Correlations between maternal attributions (using scenario measure, WAQ) and

arent role stress and satisfaction (l-tailed)

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Mothers Perception of Mothers PSI Total Stress

Parenting Performance
WAQ IO stability -.23** 34*“
WAQ IO globality -.22"'* .43***
WAQ IO locus -.09 .15+
WAQ 10 control .03 -.15+
WAQ ODD stability -.34"“"* .45***
WAQ ODD globalitL NA NA
WAQ ODD locus -.12 .01
WAQ ODD control .09 -.16*
WAQ Positive stability -.34*** .45***
WAQ Positive globality -.31*** .47""Ml
WAQ Positive locus -.12 .01
WAQ Positive control .09 -.16*

 

”“p<.001, "‘p<.01, “p<.05, +p<.1

This finding was consistent across behavior types; that is, stability and globality

ratings were related to less Optimal parent adjustment (high stress, low satisfaction) for

117

 

 

Inattentive-Overactive, Oppositional/Disobedient, and Positive/Prosocial scenario
behavior types. This finding was found consistently with the Written Analogue
Questionnaire, a measure where parents to read a standard scenario and answer as if their
child engaged in the behavior described. Results replicated in part with Recalled Incident
Interview but only when using the (PSI) Total Stress measure as an outcome indicator
(See Appendix E).

In summary, maternal attributions were related to role adjustment. Qualitatively
larger relations were found between attributions and role adj ustment when using the PSI-
SF Total Stress as the outcome indicator. In summary, when examining the relation
between attributions and adjustment regardless of child behavior or diagnostic group,
stability and globality were related to maternal adj ustment; whereas, locus and control

were not. Thus, Hypothesis 2a was supported.

Examination By Subtype

Subtype analyses revealed that attributions of locus about
Oppositional/disobedient and positive/prosocial behaviors were most consistently related
to maternal adjustment for mothers of children with ADHD-Combined type even if not
for mothers of other children. Notably, magnitude of relations between attributions and
satisfaction were comparable across subtypes; the lack of findings pertaining to ADD-

Inattentive type may be due to low sample size.

118

Table 43 Correlations between maternal attributions (using scenario measure, WAQ) and
parent role stress and satisfaction (2-tailed)

 

 

 

 

 

 

 

 

 

 

 

 

 

Controls Controls ADHD-C ADHD-C ADD- ADD-
Inattentive Inattentive
Mothers Mothers Mothers Mothers Mothers Mothers
Perception of PSI Perception PSI Total Perception PSI Total
Parenting Total of Parenting Stress of Stress
Performance Stress Performance Parenting
Perform
WAQ IO -.48** .20 -. l 7 .39“ -.29 -.05
stability
WAQ IO -.53** .41 * -.28 .35+ -. 16 .33
flbality
WAQ IO -.15 .19 -.10 .26 -.17 .09
locus
WAQ IO .20 -.16 .01 -.04 .30 .1 I
control
WAQ -.47** .20 -.25 .50” -.39 .28
ODD
stability
WAQ -. 14 .00 -.40"' .30+ .51 -.15
ODD
locus
WAQ .23 ‘ -.19 -.O6 .10 .41 .02
ODD
control -
WAQ -.47"‘* .20 -.25 .50" -.39 .28
Positive
stability
WAQ -.50*"‘ .46" -.38"I .74""MI -.42 .19
Positive
obality
WAQ -. 14 .00 .40* .30+ .51 -.15
Positive
locus
WAQ .23 -.19 -.O6 .10 .41 .02
Positive
control

 

 

 

 

 

 

 

 

 

mp<oor, "p<.01, ‘p<.05, +p<.1

119

Table 44 Correlations between maternal attributions (using scenario measure, WAQ) and
parent role stress and satisfaction (l-tailed)

 

 

 

 

 

 

 

 

 

 

 

 

 

Controls Controls ADHD-C ADHD-C ADD- ADD-
Inattentive Inattentive
Mothers Mothers Mothers Mothers Mothers Mothers
Perception of PSI Perception PSI Total Perception PSI Total
Parenting Total of Parenting Stress of Stress
Performance Stress Performance Parenting
Perform
WAQ IO -.48** .20 -.17 .39"' -.29 -.05
stability
WAQ IO -.53** .41" -.28+ .35* -. 16 .33
globality
WAQ IO -.15 .19 -.10 .26+ -.17 .09
locus
WAQ IO .20 -.16 .01 -.O4 .30 .11
control
WAQ -.47** .20 -.25+ .50" -.39 .28
ODD
stability
WAQ -.14 .00 -.4O"I .30“ .51"' -.15
ODD
locus
WAQ .23+ -. 19 -.O6 .10 .41+ .02
ODD -
control
WAQ -.47** .20 -.25+ .50" -.39 .28
Positive
stability
WAQ -.50** .46“ -.38"‘ .74*** -.42+ .19
Positive
globality
WAQ -.12 .OO -.40* .30“ .51* -.15
Positive
locus
WAQ .23+ -.19 -.O6 .10 .41+ .02
Positive
control

 

 

 

 

 

 

 

 

 

"*p<.001, “p<.01, ‘p<.05, +p<.l

120

Hypothesis 2b. Independence of maternal attributions.

This hypothesis aimed to address whether maternal attributions uniquely
predicted maternal role adjustment (over and above what child behaviors contributed).
As found in 2a, when child behaviors were not considered, mothers’ attributions of
stability and globality correlated with maternal role adjustment whereas locus and control
did not. The next step in testing for unique prediction is to correlate (or regress)
maternal attributions with maternal adjustment, controlling for child behaviors. Thus, to
test whether attributions uniquely predicted role adj ustment, independent of child
behaviors, attributions and child behaviors were entered into a regression equation
together. Type III Sum of Squares was used to examine the effect of one variable

controlling for all others entered into the equation.

Unique prediction (non shared variance)

As shown in Tables 45 and 46, maternal attributions about Inattentive/Overactive
behaviors were not related to maternal role adjustment when child behavior was also in
the model. That is, maternal attributions about child inattentive/overactive bemrvior did

not uniquely predict maternal role adjustment. In contrast, attributions about the stability

121

of mesitional/Disobedient behaviors were uniquely related to maternal adjustment,

even when child behaviors were controlled.

Table 45 Regression to test relation between maternal attributions and maternal role
adjustment independent of child behaviors (step 3) and test of moderation (step 4); with
dependent variable=PSS Satisfaction with Parentin for all models

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Predictor variable Beta Beta Beta Beta R’ R’ R’ a’
step] step2 step3 step4 Change Change Change Change
Step 1 Step 2 Step 3 Step 4
Model 1 Child lrrattn -.12 -.12 -.09 -.09 .10"
Child Hyp -.23 -.07 -.08 -.16
Child Disobedient! -.l8 -.16 -.27 .01
Oppositional
Globality about 10 -.08 -.13 .01
behavior
lrrteraction term .24 .02
hiat‘hyp'odd‘atrrib
Model 2 Child Inattn -.11 -.ll -.07 -.08 .10“
Child Hyp -.23+ -.08 -.09 -.16
Child Disobedient! -.l8 -.16 -.27 .01
Oppositional
Stability about 10 -.13 -.19+ .01
behavior
Interaction ter .25 .02
lrrat‘hyp‘ attrib
Model 3 Child Inattn -.12 -.12 -.07 -.06 .10”
Child Hyp -.23+ -.07 -.06 -.12
Child Disobedient! -.18 -.12 -.22 .01
Oppositional
Stability about ODD -.24‘ -.34" .05'
behavior
Interaction term .25+ .02+
ODD‘attnb
*"p<.001, ”p<.01, ‘p<.05, +p<.l

122

Table 46 Regression to test relation between maternal attributions and maternal role

adjustment independent of child behaviors (step 3) and test of moderation (step 4); with
dependent variable=PSI SF Total Stress for all models

 

Predictor variable Beta
step 1

Beta
step 2

Beta
step 3

Beta
step 4

R1
Change
Step 1

R1
Change
Step 2

R1
Change
Step 3

R2
Change
8&4

 

Model 1 Child Inattn .18+

.20+

.15

.15

.43W

 

Child Hyp .52‘“

.18

19

21

 

Child Disobedient/
Oppositional

.40“

.36“

.59.—

.05”

 

Globality about 10
behavior

.13

.14

.Ol

 

Interaction term

-.06

 

 

Model 2 Child Inattn .18+

.19+

.16+

.16-'-

.43m

 

Child Hyp .52m

19

 

Child Disobedient!
Oppositional

.59"

.53»

.57"

.05“

 

Stability about 10
behavior

.10

.10

.17

 

Interaction term

-.01

 

 

Model 3 Child Inattn .18-t-

.20+

.16-*-

.16+

.43m

 

Child Hyp 52""

.18

18

18

 

Child Disobedient!
Oppositional

.40“

.55"

.36"

.05“

 

Stability about ODD
behavior

.16‘

.18+

.02‘

 

 

 

 

Interaction term

 

 

 

-.03

 

 

 

 

 

mp<.001, "p<.01, *p<.05,+p<.1

An alternative to maternal attributions uniquely predicting maternal role

adjustment was the possibility that child behaviors mediated the relation between

maternal attributions and role adjustment.

 

Mothers

Mediation Model

 

 

‘attribution

 

 

 

 

Child

behaviors

 

 

 

 

A

Mothers’ role

adjustment

 

 

 

The results shown in Tables 45 and 46 suggest that child behaviors mediated the

relations between maternal attributions about child inattentive/overactive behaviors and

maternal role adjustment. flat is, the gignificant relations found between attributions

123

 

about 10 behaviors and maternal role adjustment (in 2a) became non-sigm'ficant when
child behaviors were included in the model. Child behaviors also partially mediated

relation between attributions about ODD behaviors and maternal adjustment (i.e., relation
of attribution and adjustment decreased from -.34 to -.24 using PSS and from -.45 to .16
using PSI).

To test the fill] mediation model, two additional tests were required. Specifically,
for the mediation model to be supported, not only must the prior significant relations
(between attributions and role stress) become non-significant but child behaviors must
also be significantly related to maternal attributions. Step 1 in Tables 43 and 44
displayed the significant relations between child behaviors and maternal adjustment;
significant relations were found between attributions and adj ustment only when using the
PSI Total Stress score (not when examining the parenting satisfaction score).
Additionally, as shown in Table 47, child behaviors were significantly related to maternal

attributions.

124

Table 47 Correlations between maternal attributions about child behavior in relation to
mother and teacher ratings of child behaviors (2-tailed)

 

 

 

 

 

Mom rating of child Mom rating of child Mom rating of child
Inattention hyperactivity OPP/disobedient
Behavior
WAQ IO globality .47*** .43‘” 44*"
WAQ IO stability .41"* .34‘" .35‘"
WAQ ODD stability .43‘“ .44’“ .46*‘“'

 

 

 

 

 

mp<.001, "p<.01, *p<.05, +p<.l

Thus, child behaviors appeared to mediate the relation between maternal
attributions about 10 behaviors and maternal stress (when using PSI as outcome); that is,
when child behaviors were in the model, the relation between maternal attributions and
maternal stress was no longer significant. Child behavioral severity, thus, explained the
relation between maternal attributions and maternal stress (using PSI measure but not

with PSS measure).

Additional analysis: Moderation by severity of child behavior

A third possibility was that dimensional severity of child behavior moderated the

relation between maternal attributions and role adjustment.

Moderation By Severity of Child Behaviors

 

Child
behaviors

 

 

 

 

 

Mothers Mothers’ role

 

‘attribution adjustment

 

 

 

 

 

 

125

For moderation to be supported, the interaction term (child behaviors‘attribution)
should be significant after controlling for the variance due to main effects. The test for
moderation by severity of child behaviors was also shown in Tables 45 and 46. The
interaction term was not significant. That is, the moderation effect was not found when
examining the interaction between attributions and child behavioral severity.

In summary, attributions about Oppositional/Disobedient behavior appear to
function somewhat differently than attributions Inattentive/Overactive behaviors.
Stability attributions about oppositional/disobedient behaviors predicted role adjustment
above and beyond that accounted for by child behaviors to a small but significant degree.
Child behaviors partially mediated this relation. In contrast, attributions about child
inattentive/overactive behaviors appeared to be entirely mediated by severity of child
behaviors; that is, child behaviors explained the relation between attributions about 10
behaviors and parent role adjustment.

The significant amount of shared variance among child behavior variables (see
methods section) presents some interesting challenges to the analyses and understanding
of the results (a complication to studying ADHD). In order to provide comprehensive
data so that the subtleness of the effects of such shared variance may be discussed, Tables
48 and 39 are included here, showing the test for uniqueness of attribution and test for
moderation when controlling only for the behavior related to the attribution of interest.
When controlling for only the belLvior related to the specific attribution. the moderation
effect was significant for attributions of stability about 10 behaviors and severity of child

10 behaviors but only in relation to PSS.

126

Table 48 Regression to test relation between maternal attributions and maternal role
adjustment independent of child behaviors (step 2) and test of moderation (step 3); with
dependent variable=PSS Satisfaction with parenting performance (controlling only for
behavior related to the attribution of interest)

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Beta step 1 Beta step 2 Beta step It” Change R2 Change R2 Change
3 Step 1 Step 2 Step 3

Model 1 Child Inattn -. 12 -.08 -.15 .10”

Child Hyp «.23+ -.21+ -.49‘

Globality about -.10 -.23+ .01

IO behavior

Interaction term .00+ .03+
Model 2 Child Inattn -.11 -.07 -.15 .10“

Child Hyp -. 13+ -.22+ -.52“

Stability about -.14 -.32“ .02

IO behavior

Interaction term .51"I .05”
Model 3 Child -.3 I " -.20" -.42"' .10“I

Disobedient!

Oppositional

Stability about -.26“ -.45“ .06“

ODD behavior

Interaction term .36 .02
"‘p<.001, "p<.01, l‘p<.05, +p<.l

127

 

Table 49 Regression to test relation between maternal attributions and maternal role
adjustment independent of child behaviors (step 2) and test of moderation (step 3); with
dependent variable=PSI-SF Total Stress (controlling only for behavior related to the

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

attribution of interest)

Predictor variable Beta step 1 Beta step 2 Beta step R2 Change R2 Change R2 Change

3 Step 1 Step 2 Step 3

Model 1 Child Inattn .18+ .12 .13 .43‘"

Child Hyp 52““ .49'" 54“"

Globality about .16+ .19+ .02+

10 behavior

Interaction term -.08 .00
Model 2 Child Inattn .18+ .14 .14 .43‘“

Child Hyp .52‘" 51*“ 52“"

Stability about 10 .12 .13 .01

behavior

Interaction term -.02 .00
Model 3 Child .66‘“ .57'" 55“" .43‘“

Disobedient/

Oppositional

Stability about .21“ .20 .04“

ODD behavior

Interaction term .02 .00
”‘p<.001, I"‘p<.01, l‘p<.05, +p<.l

Hypothesis 3 — Moderation by Child Diagnosis

This hypothesis aimed to address whether the relation between parent attribution and

parental role adjustment moderated by child diagnosis.

 

 

Mothers

‘attribution

Diagnosis as Moderator

 

 

Child diagnosis

 

 

 

,l

 

 

 

128

 

Mothers’ role

adj ustment

 

 

 

ADHD diagnosis predicted parental adjustment (r=.36, p=.002; see also Tables 50
and 51 beta step 2) and attributions marginally predicted parental adjustment (r ranges
from -.22 to -.34, p<01 to p<.01, refer back to Table 41).

As shown in Table 51, ADI-ID diagnosis partially moderated the relation between
maternal attributions and adjustment when using PSI Total Stress as an outcome variable
(but not when using PSS Satisfaction with Parenting Performance as outcome). Although
not hypothesized, Tables 50 and 51 show that diagnosis partially mediated (to a ma!
§x_te_n_t) the relation between maternal attributions and adjustment; that is, the magnitude
of the relation decreased when diagnosis was included in the model. Partial mediation
was found when using the PSS Satisfaction with Parenting as well as when using the PSI

Total Stress as outcome variable.

Table 50 Regression to test moderation of diagnostic variable (control vs. any type of
ADHD) in the relation of maternal attributions and role adjustment, dependent
variable=PSS satisfaction with

   
   

Predictor variable Beta step 1 Beta step 2 Beta step 3 R’ Change R’ Change a’crmge
Step 1 Step 2 Step 3
about 10 -.51 -.44
ADHD -.1
term

ahmt
ADHD
mwmamnumn

ODD
ADHD

term
mp<.001, "p<.01, “p<.05, +p<.l

129

Table 51: Regression to test moderation of diagnostic variable (control vs. any type of
ADHD) in the relation of maternal attributions and role adj ustment, dependent
variable=PSI Total Stress as outcome

Predictor variable Beta step 1 Beta step 2 Beta step 3 R’ Change R’ Change R’ Change
1 2 3

Globality about 10 .53 .34" .16 .28‘”

behavior

ADHD .44 .31

Interaction term .40“

about .05
ADHD .28"
m an

about ODD . .16
behavior
ADHD . .16
Interaction term .42

mp<.001, ”p<.01, ‘p<.05, +p<.1

 

Additional Tests to Further Examine the Role of Locus and Controllability

To be certain that effects of controllability and locus were not masked by
diagnostic or mediation effects, the moderation test was examined for controllability and
locus. None of the interactions terms were significant, indicating that possible relations
with adj ustment and attributions of locus and controllability were not masked by ADHD
vs. non ADI-ID diagnostic categorization nor by medication status. However, as
discussed, subtype analyses reveal that attributions of locus about
Oppositional/disobedient and positive/prosocial behaviors were related to maternal
adjustment for mothers of children with ADHD-Combined type even if not for mothers

of other children.

Summary of Maternal Attribution Findings
As predicted, mothers of children with ADHD attributed disruptive child
behaviors to more stable, global, and geontrollable child factors than parents of control

children. No group differences were found pertaining to attributions of locus.

130

Second, as predicted, stability and globality attributions were related to higher
levels of role stress and lower levels of role satisfaction. Contrary to expectations,
attributions pertaining to locus and controllability were not significantly related to role
adjustment.

Third, maternal attributions about child inattentive/overactive behavior M

uniquely predict maternal role adjustment. In contrast, attributions about the stabilig of

 

Omsitional/Diflredient behgviors were uniquely related to maternal adjustment, even

when child behaviors were controlled. Fourth, dimensional severity of child behaviors
mediated (but did not moderate) the relation between maternal attributions and stress.
Notably, child diagnosis partially moderated the relation between maternal attributions
and adjustment.

In summary, the relation between maternal attributions and adjustment does
appear to depend upon child behaviors. Indeed, a compelling finding was the moderation
finding; specifically, child diagnosis served as a moderator for attributions about 10
behaviors in relation to maternal role adjustment.

In order to facilitate understanding and discussion of the findings, a brief

summary of key findings is provided in Table 52.

131

Table 52 Summary of primary findings

 

 

 

Primary Question Finding Implications/relation to
literature/further questions
Attributions by group Attributions difl‘ered by diagnosis in Partial replication of Johnston, except in
direction expected, except attributions regard to positive beliefs by diagnosis
about positive behaviors not in direction
expected
Group difi‘erences in attributions of
Subtype analyses revealed that group locus did appear to be masked by
difl'erences were frequently but not medication status.
exclusively driven by differences
between ADHD-Combined subtype and
controls.
Attributions by belief Attributions also differed by belief about
about diagnosis diagnosis, including positive behaviors

in direction expected

 

Dimensional relation

Attributions were related to adjustment

This analysis did not look at attributions

 

 

 

between in direction expected, except that and adjustment by group.

attributions and globality and stability of Positive as well Controllability attributions were not

adjustment as problem behaviors were related to related to adjustment.
stress
Controllability attributions were not Subtype analyses reveal that attributions
related to adjustrrrent when groups of locus about Oppositional/disobedient
combined and positive/prosocial behaviors were

related to maternal adjustment for
mothers of children with ADD—
Combined type even if not for modrers
of other children.

Attributions about controllability did not
appeared to be obscured by combining
the two groups; that is, no sig. Effects
were found btwn controllability and
adjustment even when exarrrined
separated by subtype.

Pathways for Only attributions about ODD/disobedient Attributions about ODD behavior appear

understanding behaviors were uniquely related to to have more powerful impact on

relation between mothers’ adjustment maternal adjustment than do attributions
attributions and . . . . about 10 behavror.

“mum!“ Child behavror mediated relation btwn . . . .
mom attrib and adjustment for attrib Child behavror notably mediates relation
about IO behaviors between attributions and adjustment
Child behavior partially mediated rel
btwn rrrom attrib and adjustment for
attrib about OPP behaviors
No moderation by severity of child
behaviors and attributions (one finding
supported moderation but with attrib
about 10 stability using PSS controlling
only for 10 beh).

Attributions and Diagnosis moderated relation

Adjustment - group
effects?

 

between mothers’ attribution and
adjustment but only for 10 behaviors
using PSI

Diagnosis partially mediated relation
between mom attrib and adjustment

 

 

132

 

Covariates
No group differences were found on any covariates (See Table 1). Thus, analyses were
not conducted controlling for covariates at this point in time except for the examination
of relation between maternal attributions and maternal adjustment with aggression
controlled.

Table 53 Correlations between maternal attributions (using scenario measure, WAQ) and

parent role stress and satisfaction, with child aggression/oppositional behaviors controlled
2-tailed)

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Controls Controls ADHD-C ADHD-C ADD-Inattentive ADD-
Inattentive

Mothers Mothers Mothers Mothers Mothers Mothers
Perception of PSI Total Perception of PSI Total Perception of PSI Total
Parenting Stress Parenting Stress Parenting Stress
Performance Perfomrance Performance

WAQ IO -.49" .20 -.09 .16 -.29 -. 12

stability

WAQ IO -.50"”" .35‘ -.22 .08 -.04 .08

h_globality

WAQ IO -.09 .12 -.04 .09 -.l9 .15

locus

WAQ IO .14 -.07 .01 -.07 .44 -.14

control

WAQ -.48“ .19 -.20 .34+ -.26 -.25

ODD

stability

WAQ -.09 -.08 -.36+ .13 56+ -.24

ODD locus (p=.05)

WAQ .21 -.16 .01 -.17 .51 -.15

ODD

control

WAQ -.48" .19 -.20 .34+ -.26 -.25

Positive

stability

WAQ -.44” .38" -.35+ .55“ -.30 -.37

Positive

lobality

WAQ -.09 -.09 -.36+ .13 56+ -.24

Positive (p=.05)

locus

WAQ .21 -.16 .01 -.17 .51 -.15

Positive

control

I"""p<.001, I“p<.01, l"p<.05, +p<.l

133

Table 54 Correlations between maternal attributions (using scenario measure, WAQ) and
parent role stress and satisfaction, with child aggression/oppositional behaviors controlled

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

l-tailed)
Controls Controls ADHD-C ADHD-C ADD- ADD-
Inattentive Inattentive
Mothers Mothers PSI Mothers Mothers PSI Mothers Mothers
Perception of Total Stress Perception Total Stress Perception PSI Total
Parenting of Parenting of Parenting Stress
Performance Performance Performance
WAQ IO -.49" .20" -.O9 . I6 -.29 -.12
stability
WAQ IO -.49“ .35’ -.22 .08 -.04 .08
globality
WAQ IO locus -.09 . 12 -.04 .09 -. l9 . 15
WAQ IO control .14 -.07 .02 -.07 .44 -.I4
WAQ ODD -.48" .19 -.20 .34‘ -.26 -.25
stability
WAQ ODD -.09 -.08 -.36" .13 .56‘ -.24
locus
WAQ ODD .21 -.16 .01 -.I7 .51+ -.15
control
WAQ Positive -.28 .19 -.20 .34‘ -.26 -.25
stability
WAQ Positive -.44'"I .26‘ -.35‘ .55” -.30 c.38
_globality
WAQ Positive -.09 -.08 -.36"' .13 .56“ -.24
locus
WAQ Positive .21 -.16 .01 -. l 7 .5 1+ -. 15
control
*”p<.001, “p<.01, ‘p<.05, +p<.l

Exploratory Analyses 1 : Maternal Attributions and child behaviors -

This analysis tested the relation between maternal attributions in relation to
severity of child behavior (collapsed across diagnostic groups). The predication was that
maternal attributions that disruptive child behaviors are internal, uncontrollable, stable,
and global would correlate with greater severity of dimensional child disruptive
behaviors (inattention, hyperactivity, and aggression), whether using teacher or parent
ratings of child behaviors.

This exploratory analysis was conducted even though the author had anticipated
(and found) that attributions would differ by group. Thus, if attributions differed by

group, dimensional relations would expected be obscured on those domains for which

134

differences were anticipated. Notably, in testing the possibility that child behaviors
might mediate relation between parent attributions and adjustment, the relations between
mothers’ attributions and child behaviors was tested (Table 55), this was only for
particular attributions. .

In order to examine dimensional relations between maternal attributions and
severity of child behavior, correlations were conducted (see Table 55). As shown in
Table 55 maternal attributions about stability and globality were consistently positively
correlated with severity of child behavior problems, across types of behaviors. Also,
corresponding with predictions, maternal attributions of control were negatively
correlation with child behavior problem severity. This result was found with child
ADHD behaviors (both Inattention [10 control and inattention, r=.29, p<.001) and
Hyperactivity (IO control and hyperactivity, r=-.20, p<.05]; see Table 55 but with not
child Oppositional/Disobedient behaviors (OPP control and OPP behaviors, r=.01, n.s.).
Also, as shown in Table 55 these results were found when using both mother and teacher
ratings of child behaviors. As shown in Appendix F, results were similar although less
strong (i.e., correlations were smaller in magnitude and with lower p values) when using

the Recalled Incident Interview measure.

135

Table 55 Correlations between maternal attributions about child behavior in relation to

mother and teacher ratings of child behaviors (3-factor analysis) (2-tailed)

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Mom rating Morn rating of Mom rating of Teacher. Teacher rating Teacher rating
of child child child rating of of child of child
inattention hyperactivity OPP/disobedient child Hyperactivity OPP/disobedient
Behavior infatuation BeLavior
WAQ IO .41‘" .34‘" .35‘" .27” .28" .17+
stability
WAQ i0 .47“. .43‘“ .44‘" .34‘“ .25“ .23‘
_glpbality
WAQ 10 .08 .08 .15 .09 .09 -.0l
locus
WAQ 10 49"" -.20‘I -. 10 -.20‘ -.23‘ .01
control
WAQ ODD .43‘” .44‘” .46‘“ .25‘ 32““ .30“
stability
WAQ OPP -.04 .01 .05 .07 .07 .03
locus
WAQ OPP -.3 l ‘“ -.lS .Ol -.20" -.18+ -.07
control
More WAQ .43‘“ .44‘" .46‘” .25‘ .32‘” .30”
Positive
stability
WAQ .47'” .49". .52‘“ .32" .34‘“ .34‘“
Positive
_globality
WAQ -.04 .Ol .05 .07 .07 .03
Positive
locus
WAQ -.32“‘ -.lS .Ol -.20‘ -.18 -.07
Positive
control
mp<.oor, ”p<.01, *p<.os,+p<.1

Exploratory Analysis 2: Maternal responsibility and control

In prior studies, attributions about child controllability have been found as key to
parental adjustment (Dix & Grusec, 1985). The extent to which parents feel responsible
for their child’s behavior and the extent to which they feel both responsible and able (or
unable) to control their child’s behavior might also be important in their adjustment. An
additional exploratory analysis was that parental experience of responsibility might
interact with perceived ability to control child’s behavior in predicting parental role
adjustment. [As an exploratory analysis, attributions about prosocial behavior were

included].

136

As shown in Table 56, maternal attributions about own responsibility and ability
to control child’s behavior were related to role adjustment when considering child
Oppositional/Defiant but not Inattentive-Overactive behaviors. Correlations were more
consistently found between maternal attributions about own role and Satisfaction with
Parenting Performance than with PSI Total stress. The magnitude of significant
correlations ranged from -.19 (p<.05) to -.39 (p<.001). Notably, the Hoza interaction
measure that measured mothers’ attributions about own role in relation to child

compliance and non-compliance appeared to be a good measure for this analysis (See

Table 56).

137

Table 56 Correlations between attributions about parent’s own role in child behavior
(attributions of responsibility and control) in relation to parents’ role adjustment (PSS
satisfaction and PSI Total Stress). (2-tailed)

 

PSS Satisfaction with Parenting
Performance

PSI Total Stress

 

WAQ Parent responsibility to
control child behavior
Inattentive-Overactive (IO)
behavior

.1)6

-.01

 

WAQ Parent ability to control child
behavior
10 behavior

.22

 

 

WAQ Parent responsibility to
control child behavior
Oppositional (OPP) behavior

-.l9‘

.07

 

WAQ Parent ability to control child
behavior
Oppositional (OE) behavior

.15

-.19

 

 

WAQ Parent responsibility to
control child behavior
Positive (POS) behavior

.12

-.lO

 

 

RII Parent responsibility to control
child behavior
inattentive-Overactive (IO)
behavior

-.09

-.03

 

RiI Parent ability to control child
behavior
10 behavior

-.23

-3l“

 

 

RII Parent responsibility to control
child behavior
Oppositional (OPP) behavior

-.20“

-.12

 

RII Parent ability to control child
behavior
Oppositional (OPP) behavior

.39“I

-.37“

 

 

RII Parent responsibility to control
child behavior
Positive (POS) behavior

.15

-.12

 

Rii Parent ability to control child
behavior
Positive (POS) behavior

-.06

.ll

 

 

Hoza good parenting and ability to
control
Child non-compliant

.16+

./1.

 

Hoza ability to control child
compliant behavior

-.36$i¢

.19"

 

Hoza effort to control
Child non-compliant and compliant
behavior

.20“

-.il

 

 

General good parenting
Child non-compliant and compliant
behavior

 

-.l9"

 

 

mp<.001, ”p<.01, ‘p<.05, +p<.l

138

 

For the moderation analyses, only the PSS Satisfaction with parenting
performance measure was used. Both the WAQ and RH were used as attribution
measures. The RII was included as it was more reliably related to role adjustment
compared to the WAQ. The Hoza measure did not correspond to the exploratory
hypotheses (whereas Johnston’s WAQ and RI] did).

The test for the interaction between maternal attributions of responsibility and
ability to control child Oppositional/Disobedient behavior was not significant, indicating
that these two attributions to not interact with each other to predict maternal role
satisfaction. Notably, however, maternal attributions about ability to control child
Oppositional/Disobedient behaviors uniquely predicted maternal role satisfaction above
and beyond attributions of responsibility. Thus, mothers’ attributions about ability were
more related to role satisfaction than responsibility.

As shown in Table 57, maternal attributions about ability to control child
oppositional/disobedient behaviors were predictive above and beyond that which was

predicted by mothers’ ratings of child oppositional/disobedient behavior.

139

Table 57 Moderation test controlling for child behavior, using RII; PSS as outcome

 

 

 

 

variable
Modeil
Predictor variable Beta step Beta step Beta step R2 Change R2 Change R2 Change
1 2 3 Step 1 Step 2 Step 3
R11 Attribution of -. 19 -.20 .00 .19”
parental responsibility
OPP behavior
RII Attribution of .38" .33"I .44“
parent ability to
control of child OPP
behavior
Mother rated Child -.26" ..03 .06“
Oppositional/Disobedi
ent behavior
interaction term -.32 .01

 

 

 

 

 

 

 

 

 

 

_c_hild’§ behgvior were related to role adjustment when considering child

In summgy, maternal attributions about own responsibility and abilig to control

mesitional/Defiant but not hgttentive-Ovegctive behgviors. Maternal attributions

about ability to control child Oppositional/Disobedient behaviors uniquely predicted

maternal role satisfaction above and beyond attributions of responsibility. Thus,

mothers’ attributions about ability were more related to role satisfaction than attributions

about responsibility.

140

 

Exploratory Analysis 3: Mediation by maternal attributions
Child behaviors were examined as a mediator of the relation between child behavior and

parental role adjustment.

 
  
 

   
  

   
    
 

Child
Behavior

Maternal
Attributions

Maternal
Adjustment

There are four components to testing the mediation model. First, are child
behaviors related to maternal adjustment (exploratory analysis 3a)? As shown in Table
58, severity of child behavior was significantly related to parent role stress.

Table 5 8 Correlations between child behaviors and maternal role adjustment (2-tailed)

 

 

 

 

 

 

 

 

 

Satisfaction with parenting Parent Stress Total
Mother rating of child -.27** 55".
inattention
Mom rating of child -31 «in .65*“
Hyperactivity
Mom rating of child -,31*** . 67*"
@positional behavior
Teacher rating of child -. 15 .40":
inattention
Teacher rating of child ”34*" .51":
Hyperactivity
Teacher rating of child -.31** .49":
Oppositional behavior

 

 

 

 

m p<.001, ** p<.01, *p<.05, +p<.l

The second step of testing the mediation model was to test whether child behavior
is related to maternal attributions. As shown in exploratory analysis la, child behavior
was significantly related to maternal adjustment (such that more severe behavior is
correlated to poor role adjustment). Specifically, attributions of globality and stability

and uncontrollability were positively related to poor role adjustment.

141

The third step was to examine whether maternal attributions are related to
maternal adjustment (hypothesis 2). This was tested in hypothesis 2. The results were
that greater stability and globality were found to be associated with greater maternal role
stress and lower maternal role satisfaction (refer back to Table 41). The findings were
strongest (size of correlation and p-value) when using the PSI Total Stress score as
outcome measure. No significant relations were found between attributions of
controllability and locus and role adjustment.

The final step (exploratory analysis 3b) in testing mediation was to examine
whether the relation between child behavior and maternal adjustment became non-
significant (full mediation) or became less significant with a decrease in the magnitude of
the correlation (partial mediation). Because the findings in the first three steps were
strongest when using the PSI as an outcome measure and the WAQ as attribution
measure, the mediation model was tested using only the WAQ to measure globality and
stability predicting role adjustment as measured by the PSI Total Stress score.

As shown in Table 59, attributions were not found to mediate the relation between
child behavior and maternal role adjustment. Although the magnitude of the correlation
between child behavior and role adjustment decreased slightly, the attribution itself was

not significant at step two (see model 2).

142

Table 59 Regression to test mediation of child behavior and maternal role adjustment by
maternal attributions with dependent variable=PSI-SF Total Stress

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Predictor variable Beta step 1 Beta step 2 Beta step 3 R1 Change R’ Change R’ Change
Step 1 Step2 Step 3

Model 1 Child inattn .20‘ .16+ .16 50’”

Child Hyp .l7 .17 .20

Child Disobedient! .42‘“ .39” .42”

Oppositional

Globality about 10 .02 .13 .01

behavior

Interaction term -.08 .00
Model 2 Child inattn .20‘ .18+ .18+ .50‘”

Child Hyp .17 .17 .17

Child Disobedient! .42‘“ .40‘“ .40”

Oppositional

Stability about 10 .07 .06 .00

behavior

Interaction term .01 .00
Model 3 Child lnattrr .20‘ .16+ .16+ .50‘“

Child Hyp .17 .17 .18

Child Disobedient! .42‘” .37“ .39”

Oppositional

Stability about .14+ .15+ .01

ODD behavior

interaction term -.04 .00

mp<.oor, ”p<.01, *p<.05, +p<.l

Results controlling for only behavior corresponding to attribution are shown in
Appendix G. When only controlling for child behavior corresponding to attribution of
interest, there appears to be a slight mediation effect. However, as shown fiom results in
Table 59, the variance under consideration is better explained by co-occurring child
behaviors. [Note that as shown in Tables 50 and 51 even when reaching diagnostic

levels, diagnosis did partially mediate the relation between maternal attributions and
adjustment]

143

CHAPTER 5
RESULTS FOR STUDY 2

Child Results

Exploratory Analysis 4: Child attributions

Child attributions about their own disruptive behaviors were expected to differ by
diagnostic status and be correlated with severity of own behaviors
Exploratory Analysis 4a: Drfl'erences by diagnostic group
Children with ADHD made attributions about Oppositional and Prosocial
behaviors that were more controllable compared to children without ADHD. They also
attributed Inattentive/Overactive behaviors to more internal causes.

Table 60 Test of differences in attributions made by children with ADHD diagnosis
versus children without a behavioral disorder, using the WAQ and Hoza measures

 

 

 

 

 

 

 

 

 

 

 

Control ADHD any F statistic (dt) P value Value on attrib Attributions
type measure about ADHD
group in
comparison to
control
WAQ [O 2.3 3.4 F(1,78)=.3O n.s. l=specific
_globality (. 9) (l .0) 5=global
WAQ 10 2.7 3.1 F(l,78)=5.0 P<.05 l=extemal ADHD more
locus (. 8) (. 8) S=intemal internal
WAQ IO 3.4 3.3 F(l,78)=l .5 n.s. l=uncontrollable
control (.5) (.6) 5=controllable
WAQ OPP 2.0 2.2 F(l,78)=.7 n.s. l=specific
_globality (l .0) (l .2) 5=global
WAQ OPP 3.4 3.2 F(l,78)=.3 n.s. l=extcrnal
locus (l .0) (l .2) 5=internal
WAQ OPP 4.2 3.7 F( l ,78)=5.3 P<.05 l=uncontrollable ADHD less
control (. 7) (l . l) S=controllable controllable
WAQ 3.9 3.5 F(l ,78)=2.8 P<. l. l=specific
Positive (1.1) (1.1) 5=global
lobality
WAQ 3.8 3.9 F(l,78)=.37 n.s. l=extemal
Positive (.9) (.9) 5=internal
locus
WAQ 4.4 3.7 F(l ,78)=9.0 P<.Ol l=uncontrollable ADHD less
Positive (.9) (1.1) 5=controllable controllable
control

 

 

 

 

 

 

 

 

*IO=Inattentive-Overactive type of behavior
*OPP=oppositional/disobedient type of behavior

144

Exploratory Prediction 4b: Child attributions and adult ratings of behavior.

Child attribution ratings of extemality, stability, globality, and uncontrollability
were expected to be related with greater severity of adult-rated child hyperactivity,
inattention, and aggression (collapsed across child diagnostic groups). As shown in
Table 61, child attributions were significantly related to mother and teachers’
dimensional ratings of severity of child disruptive behaviors.

Table 61 Correlations between child attributions about child behavior in relation to

 

 

 

 

 

 

 

 

 

 

mother and teacher ratings of child behaviors (3 -factor analysis) (2-tai1ed
Mom Mom rating Mom rating of Teacher Teacher Teacher rating of
rating of of child child rating of rating of child child
child hyperactivity OPP/disobedient child Hyperactivity OPP/disobedient
inattention Behavior inattention Behavior

WAQ 10 .18+ .03 .03 .10 .08 .ll

hQibality
WAQ IO .22 .27“I .22‘ .21“ .17 .ll
locus
WAQ IO -.28" -.21" -. 16 -.20+ -.20‘ -.18+
control
WAQ .26“ .18" .16+ .03 .16 .09
OPP

maria
WAQ -. 13 -.05 .Ol -.09 -.l l .02
OPP
locus
WAQ -.26“ -.20* -.ll «23‘ -.26* -.18+
OPP '
control
WAQ .13 .01 .Ol -.09 -. 13 -. 19
Positive

lobality
WAQ .01 .09 .02 -.05 .Ol .01
Positive
locus
WAQ -.33""" -.08 .00 -.24‘ -.07 .00
Positive
control

 

 

 

 

 

 

 

 

 

*"p<.001, "p<.01, ‘p<.05, +p<.1

145

Table 62: Correlations between child attributions about child behavior in relation to

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

mother and teacher ratin is of child behaviors (3 -factor analysis) (1 -tailed)
Mom rating Mom rating Mom rating of Teacher rating Teacher rating Teacher rating of
of child of child child of child of child child
Inattention hyperactivity OPP/disobedient inattention Hyperactivity OPP/disobedient
Behavior Behavior
WAQ lo .18‘ .03 .03 .10 .08 .11
_giobality
WAQ lO locus .22” .27" .22" .21- .l7- .11
WAQ 10 control -.28“ -.21‘ -.16‘ -.2o‘ -20. his.
WAQ OPP .26" .18" .16‘ .03 .16+ .09
+_gl:obality
WAQ OPP locus -.13+ -.05 .01 -.09 -.ll .02
WAQ OPP -.26” -.20. -.ll -23. -.26“ -.18‘
control
WAQ Positive .03 .01 .or -.09 -.13 -.l9‘
globality
WAQ Positive .01 .09 .02 -.05 .or .01
locus
WAQ Positive -.33m ..08 .oo .24" -.07 .00
control
mp<.001, "p<.01, ‘p<.05, +p<.l
Subtype Comparisons

When testing for group differences considering ADHD subtype, a significant

result was found for child attributions of locus pertaining to oppositional/disobedient

behaviors and for control pertaining to Oppositional behavior (see Table 63).

Specifically, children with ADHD-Combined type made attributions that

Inattentive/Overactive behaviors were more internal compared to their peers without

ADI-ID. Additionally, children with ADHD-Combined type attribute

oppositional/disobedient behaviors to less controllable causes than children with ADD-

Inattentive type or than children without ADHD.

146

 

Table 63 Results of child attributions, testing for group differences when comparing the
independent variable of child diagnosis with four levels (ADHD-C, ADD, subthreshold,

and controls), dependent variable=attribution domain by behavior type

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

a h C
Control ADHD- ADD - Sub- F stat p- Value on Conclusions
Combined inattentive threshold (di) value attrib
ADHD measure
WAQ 10 2.3 2.4 2.3 2.2 F(3,109) n.s. l=specific
_glfobality (9) (l .0) (.9) (.9) -.20 We]
WAQ 10 2.7‘ 3.3' 2.7 2.9 F(3,lO9) P<.01 l-extemal ADHD-C
locus (.8) (.7) (.8) (.6) =42 Stintemal attribution
behaviors to
more internal
factors
Differ btwn
ADHD-C and
add was p-.055
WAQ 10 3.4 3.3 3.3 3.2 F(3,109) n.s. l-uncontlolla
control (.5) (.6) (.5) (.6) -l.1 ble
S-controllable
WAQ 2.0 2.3 2.0 2.1 F(3,109) n.s. l-specific
OPP (1.0) (1.3) (.9) (1.1) -.7 S-global
it:
WAQ 3.4 3.1 3.5 3.5 F(3,109) n.s. l=external
OPP (1 .0) (1 .3) (1 .0) (.8) -.7 5=intemal
locus
WAQ 42 7 3.5‘ 4.4‘ 3.9 F(3,109) P<.01 l-uncontrolla ADHD-C
OPP (.7) (1 .1) (.7) (.9) 134.8 ble attributed
control S-controllable behaviors to
less
controllable
causes than
controls or
ADD children
WAQ 3.9 3.5 3.7 3.7 F(3,109) n.s. l-specific
Positive (1.1) (1.1) (.9) (1.3) -.9 S=globa1
1‘25”“)
WAQ 3.8 4.0 3.9 3.8 F(3.109) n.s. lIextemal
Positive (.9) (.9) (.9) (.8) -.2 5=internal
locus
WAQ 4.4 3.7 3.5 3.8 F(3,109) P<.05 l=uncontrolla ADHD-C and
Positive (.9) (1.0) (1.5) (1.2) -3.1 (p-.O ble controls p-.07
control 28) S-controllable ADHD-C and
ADD p-.099

 

 

‘iO-inattentive-Overactive type of behavior; ‘OPP-oppositional/disobedient type of behavior
a indicates significant difference between control and adhd combined poops

b indicates siplificant difference between control and add inattentive poops

c indicates siprificant difference between control and “subthreshol ” poops

d indicates significant difference between adhd and add poup

e indicates siprificant differences between add and subthreshold poops

f indicate siprificant difference between adhd and “subthreshold poops

147

 

Table 64 Correlations between mother and child attributions about child behaviors

 

Child
WAQ IO
globality

Child
WAQ
IO

locus

Child
WAQ
10
control

Child
WAQ
OPP
globality

Child
WAQ
OPP

locus

Child
WAQ
OPP
control

Child
WAQ

Positive
globality

Child
WAQ
Positive
locus

Child
WAQ
Positive
control

 

Maternal
WAQ iO
stability

.15

.18+

.02

.20"I

-.04

-ll

.05

-.01

-.18

 

Maternal
WAQ iO
__globality

.08

.16+

-.06

.19“

-.07

-.03

.06

.14

-.12

 

Maternal
WAQ iO
locus

-.05

.08

-.06

-.ll

.05

.01

.06

.03

 

Maternal
WAQ 10
control

.03

-.12

.15

-.03

.12

 

Maternal
WAQ
OPP
stability

.12

.19‘

.01

.15

-.l4

-.14

-.03

-.10

 

Maternal
WAQ
OPP locus

.05

.10

.05

-.07

.16+

.21“

.01

 

Maternal
WAQ
OPP
control

.11

.05

-.07

.18+

.05

.03

.17

 

Maternal
WAQ
Positive
stability

.12

.19"I

.Ol

.14

-.l4

-.14

-.03

-.10

 

Maternal
WAQ
Positive

.16+

.10

.22‘I

-.16

-.02

.10

-.15

 

globality
Maternal
WAQ

Positive
locus

.05

.10

.05

-.07

.16+

.22‘

.01

.10

 

Maternal
WAQ
Positive
control

 

 

 

.ll

 

.03

 

 

.08

 

.18+

 

.05

 

.03

 

.17

 

148

 

CHAPTER 6
DISCUSSION

The current study had three over-arching objectives: (1) extend knowledge
regarding maternal attributions about child disruptive behaviors in an ADHD population,
(2) examine specific maternal attributions about children’s disruptive behaviors in
relation to maternal role adjustment, (including examining possible pathways of which
attributions may play a role in maternal adjustment), and (3) explore child attributions
about disruptive child behaviors. The child portion of the study was exploratory. in fact,
one of the endeavors of assessing child attributions was to create a child version of
Johnston’s (Johnston et al., 2000; Johnston & Freeman, 1997) Written Analogue
Questionnaire. Results pertaining to child attributions are discussed last.

Several points are worth noting about the findings reported here. First, a number
of findings fiom the initial studies of attributions made by mothers of children with
ADHD were replicated, suggesting the robustness of these effects, whereas other key past
findings did not replicate. In particular, attributions of stability and globality consistently
differed by diagnostic group, replicating prior study findings. Secondly, maternal
attributions were related to maternal adjustment, with stability about disobedient
behaviors emerging as an important factor even when child behaviors were controlled.
Thirdly, the relation between maternal attributions and their adjustment was moderated at
least partially by child diagnosis when considering attributions about child inattentive-
overactive but not disobedient behaviors. These main findings will be discussed in turn,
before considering the child findings and limitations of the study.

Understanding the differences in attributions made by mothers of children with

ADHD versus mothers of children without ADHD lays the groundwork for examining

149

the different possible functions that maternal attributions about child behaviors serve in
relation to maternal adjustment. If, as the current findings suggest, mothers of children
with ADHD make attributions about their children’s behaviors that differ fiom those of
other parents, these attributions may have implications for maternal affective and
behavioral responses to their children’s behavior (Dix & Grusec, 1985; Weiner, 1980).
The implications for the effects of maternal attributions on their adjustment are multifold.
For example, mother’s beliefs and attributions may affect the standards and challenges
that they present to their children as well as their expectations and consequential support
or encouragement to perform relevant behaviors. Learned helplessness is characterized
both by attributions of stability, globality, intemality, and uncontrollability and by
depressed affect and relinquishment of purposely effective activity (Peterson, 1988). The
pattern of attributions8 made by mothers of children with ADHD mirrors this depressive
or “pessimistic attributional style.”

A pessimistic attributional style contrasts with “positive refi'aming,” a cognitive
way of thinking about problems that is associated with positive adjustment to problems in
individuals (Peterson, 1988) and in families (McCubbin & Patterson, 1983; Podolski &
Nigg, 2001). Intriguingly, one researcher has found some evidence to indicate that the
pattern of attributions normally associated with poor adjustment may be associated with

better adjustment in children with ADHD (Milich & Okazaki, 1991). It is similarly

 

' footnote: Literature pertaining to learned helplessness considered the individual’s attributions of intemality
and controllability in reference to that individual. In the current study, attributions made by mothers pertained to
whether or not the mother saw that child’s behavior as internal to the child or within the child’s control rather than

within the mother’s control.

150

possible that the pattern of attributions normally associated with poor adjustment may
serve a different function in mothers of children with ADHD.

Notably, in initial studies the pattern of attributions made by mothers of children
with ADHD differed fiom those attributions made by mothers of children without ADHD
(Johnston & Freeman, 1997 ; Sobol, et al., 1989). Interestingly, this pattern differed floor
that which has been related to poor parental adjustment. This posed a curious conundrum
for researchers of parent attributions and adjustment because parents of children with
ADHD consistently report higher levels of distress compared to their counterparts. Do
attributions operate differently in parents who have a child with ADHD 9;; were the initial
findings pertaining to differences between mothers of children with and without ADHD
unstable or unreliable findings? Replication of the group differences was thus clearly in

order.

Replication of Group Effects

Maternal attributions have recently been suggested as an important process and
family context variable in families where a child has ADHD (Johnston & Mash, 2001;
Wright et al., 2000). Despite long-standing support for the importance of attributions in
the general coping and social cognitive literatures, few studies have examined attributions
in families where a member has ADHD. However, two laboratories: one in Canada
(Johnston & Freeman, 1997; Johnston & Mash, 2001) and one in the US. (Hoza et al.,
2000) have begun to explore maternal attributions in child ADHD. Johnston’s work
pointed to the importance of examining relations between maternal attributions and

maternal adjustment in parents of children with ADHD. Hoza and colleagues (2000)

151

provided an initial look into testing that relation although it was not a focal point of her
work.

Because maternal attributions about children’s behaviors have only recently been
studied in ADHD samples, the initial findings are at a stage where replication is crucial.
Thus, the first analyses of the current study attempted to replicate those findings.
Replicating Johnston’s findings (Johnston & Freeman, 1997; Johnston, Reynolds,
Freeman, & Geller, 1998), the current study found that mothers of children with ADHD
(any subtype) made more stable, global, and uncontrollable attributions about child
disruptive behaviors compared to mothers of children without ADHD. The replication
of this finding is significant and meaningful.

Attributions that child disruptive behaviors are stable, global, and uncontrollable
are consistent with the diagnosis of ADI-II). This consistency with criteria for ADHD
indicates a concordance between maternal attributions, diagnostic criteria, and child
behavior. Diagnostic criteria for ADHD requires that the child’s disruptive behavior be
present for at least 6 months, a criterion consistent with the attribution of stability. To
meet criteria for ADHD, a child’s behavior must also occur across setting; a criterion
consistent with attributions of globality. Also, the degree of behavior must be
“maladaptive and inconsistent with developmental level,” a factor which indicates that
the child’s behavior is beyond normal levels and uncontrollable by normal parenting
behaviors. Thus, the pattern of attributions endorsed by mothers of children with ADHD
is consistent with diagnostic criteria for ADHD. Further, the attributions are associated

with increases in child disruptive behaviors as evidenced by correlations between

152

maternal attributions and both mother and teacher ratings of child behavior. Thus,
mothers’ attributions appear consistent with the realities of parenting a child with ADHD.

This consistency between attributions made by mothers of children with ADHD
and the realities of the disorder might arguably be explained by the fact that mothers
know their child’s diagnosis. That is, while these attributions are consistent with a
diagnosis of ADHD, one might wonder if the diagnosis itself causally determined
maternal attributions. Certainly most children in the current study had a prior diagnosis.
Knowledge about ADHD and ADHD diagnosis certainly would be expected to affect
maternal attributions; however, the role of this in the current study was not directly
assessed. Although this issue was not a direct target of the study, group differences in
maternal 131% about their child’s ADHD diagnosis were examined. Twenty-nine
mothers had a belief about their child’s diagnosis that differed fi'om the actual child
diagnosis established by the assessment protocol in this study. Results differed only
slightly when examining beliefs compared to results when grouping mothers based on
“actual” child diagnosis established in the study. Thus, beliefs about diagnosis
apparently played at most a small role in the effects noted here. Yet, clearly this issues
warrants more extended follow up investigation.

Additionally, one might argue that child’s actual behavior accounted fully for
attribution results; however, the analyses that examined attributions in relation to
maternal adjustment suggest that attributions play an important role in these families even
when controlling for actual child behavior (here, as rated by mothers and by teachers).
For example, mediation analyses revealed that maternal stability and globality

attributions about oppositional behaviors were linked to maternal adjustment even when

153

controlling for child behavior. Additionally, moderation analyses revealed that for
stability attributions about inattentive/overactive behaviors were differentially related to
maternal adjustment for mothers of children with and without ADHD. Attributions of
stability were related to maternal stress for mothers of children with ADHD but not for
mothers of children without ADHD. These data suggest attributions play an important
role in maternal adjustment.

Notably, the primary results in the current study pertaining to controllability were
maternal attributions about whether or not the child could control his/her own behavior.
Some additional exploratory analyses indicated that maternal attributions about her own
responsibility and ability to control her child’s behavior also differed by diagnostic
group. The reliability of these findings warrants replication; effects in the current study
were small. This was not a focus of the current study so no conclusions are drawn here
other than to note that consideration of both child and parental control of the child’s
behavior is worth study. Attributions that the child’s behavior is uncontrollable by child
and by parent are consistent with the diagnosis of ADI-ID. Implications of such
attributions are complicated as will be discussed further in section on implications of the
findings.

One finding that was in contrast to Johnston’s research (Johnston & Freeman,
1997) and to hypotheses was that no significant group differences were found on the
domain of locus in the omnibus analyses for ADHD combined type. (Two significant
differences were found on locus when analyses data by subtype but this was a difference
between ADD-inattentive type and subthreshold groups, to be discussed later in section

on subtype findings). Thus, the current study did not replicate Johnston’s work that

154

mothers of children with ADHD differed in their locus attributions compared to mothers
of non-disordered childreng.

Although Johnston (Johnston & Freeman, 1997) found maternal attributions of
locus differed by child diagnosis, even in her studies, the locus finding was not consistent
across measures. Indeed, the locus dimension may be one of the more difficult
attribution judgments for mothers to make. If mothers ascribe to a biogenetic view of
ADHD, one would expect attributions of internal locus as Johnston has found. However,
internal locus attributions in some ways might appear to “blame” the child. Given the
uncontrollable nature of ADHD, the internal locus attribution may be challenging for
some parents to make. One would expect that if mothers ascribe to a biogenetic view of
ADHD that they would attribute the causes of their child’s Inattentive-Overactive
behaviors as internal and as inconsistent with external locus or environmental
explanations of the behavior. Two possibilities are that mothers are attempting to avoid
blaming their child for this biogenetic disorder or that the parents are thinking about the
inevitable variability across situations even while the child’s behavior is largely internally
caused. That is, even children with ADHD show variability in their behavior across
activities, suggesting some interaction with environment. These interpretations are
purely speculative. The lack of the locus finding across measures is puzzling.

A possible explanation for the null finding was that medication status was
masking an effect. Johnston and colleagues (2000) had found that maternal attributions
for child behaviors when the child was medicated were opposite of those attributions

made when the child was not medicated. Given that finding, a check on the current study

 

9 Johnston & Freeman (1997) found the locus effect using the RH but not with the WAQ. The locus effect
was not significant in the current study with either the WAQ or RH.

155

findings was to control for medication (in case medication status had cancelled out an
efi‘ect). However, even with medication status controlled, the current study did not find a
difference in locus attributions made by mothers of children with and without ADI-ID.
Although no effect was found when controlling for medication status, it is worth noting
that when children take medication, medication serves as an external substance that
controls the internal biological causal components of the child’s ADHD. Thus, if parents
consider such possible effects of medication, they might attribute remaining disruptive
behaviors (not controlled by the medication) to external factors in the environment.
Given the prevailing biogenetic view of ADHD and predominant treatment by
medication, the complexity of these issues is important to note when considering
attributions about ADHD. These issues may have contributed to the null locus findings
in the current study. In the current study, a primary and likely explanation for the null
finding is measurement error. Notably, the reliabilities of the locus measures in the
current study were poor. Further examination of locus pertaining to child ADHD
behaviors along with study on measurement of locus attributions is warranted before
conclusions can be drawn.

Issues pertaining to child medication, beliefs about medication, and maternal
attributions were not the focus of the current study but nonetheless warrant noting.
Indeed, early studies pertaining to attributions in relation to ADHD revolved around
beliefs about medication (see review by Henker & Whalen, 1991). In recent studies
(Johnston, Fine, Weiss, Weiss, et al., 2000), maternal attributions about their child’s
disruptive behaviors indeed differed when mothers rated their child’s behavior when the

child was on versus off medication. When on medication, behaviors were rated as more

156

externally controlled, possibly due to the medication as an external factor but also that the
medication may have controlled any internal factors, leaving external factors to
contribute to any disruptive behavior while the child was on medication. Certainly,
medication status contributes to maternal attributions. Even though no effect for
medication status was found in the current sample, additional research in this area is
warranted. Another consideration is whether maternal attributions affect maternal
decisions regarding whether or not to use medication. Again, the current study did not
address this but further studies might illuminate further the multiple relations between
medication status, child behaviors, parent attributions, treatment decisions, and parents’
affective and behavioral responses to a child’s ADHD.

A finding which differed fi-om what was expected but which provides some room
for speculation is that when considering differences in attributions made by mothers of
children with and without ADHD, the poop differences were found in relation not only
’ to children’s ADI-ID behaviors but also in relation to children’s Oppositional and
Disobedient and Positive/Prosocial behaviors. The similarity in maternal ratings of
children’s Oppositional/Disobedient and ADHD behaviors could be important with
implications for parent stress and adjustment. Children with ADHD often but not always
exhibit co—morbid disruptive behaviors. While impulse control and sensation seeking
(biogenetic and temperament factors) likely play an important role in this link (W aldman,
Rhee, Levy, & Hay, 2001), maternal adjustment and parenting response also are likely
important factors (as suggested by twin studies indicating that shared enviromnent effects
contribute to comorbid presentations (Rhee, Waldman, Hay, & Levy, 1999; Waldman,

Rhee, Levy, & Hay, 2001)). As outlined in the literature review, maternal stress has been

157

causally linked to the development of aggression in children (related to child oppositional
behaviors; Patterson, 1996) and has been linked to the exacerbation and maintenance of
ADHD behaviors (Hinsbaw, 1994). The similarity in rating of child
oppositional/disobedient and ADHD behavior may be one important mechanism by
which maternal stress occurs and co-morbid child aggression is promoted. Maternal
ratings of child ADHD behaviors may be influenced by their ratings of aggressive
behaviors (a negative halo effect) or ADHD behaviors may be construed as
oppositional/disobedient — a set of behaviors usually associated with intentionality and
control. Similarly, children with comorbid problems could be driving a more despairing
pattern of attributions by which the attribution style is linked to poor adjustment in
mothers.

In order to consider effects of aggressive behaviors, analyses were run controlling
for aggressive behavior. When controlling for such comorbid child problems, only the
controllability rating was different for mothers of children with and without ADHD. This
suggests that the controllability of child behaviors is indeed seen differently by mothers
of children with ADHD versus without ADHD and that this difference is not driven by
comorbid child aggressive and oppositional behaviors or maternal ratings of such
comorbid behavior. When controlling for child aggressive behaviors, group differences
between maternal ratings of stability and globality of ADHD behaviors did not remain
significant. Thus, maternal ratings of stability and globablity of ADI-[D and
noncompliant behaviors may be confounded with each other. Other research has
indicated that parents and teachers often rate children at exhibiting hyperactive behaviors

if they child exhibits aggressive behavior (Abikoff, Courtney, Pelham, & Koplewicz,

158

1993; Schachar, Sandberg, & Rutter, 1986). The current study results indicate that
differences in maternal ratings of stability and globality are affected by co-occurring
child aggressive behavior.

Attributions of stability and globality of ADHD behaviors are consistent with
diagnostic criteria for ADHD. Their lack of independence when considering comorbid
aggressive/oppositional behaviors indicates that understanding attributions about ADHD
should not be examined without considering the effects of co-occurring aggression.
Notably, childhood aggression is highly correlated with child ADHD and is associated
with poor child outcome (Barkely, 1998) as well as with maternal stress (Podolski &
Nigg, 2001). IfADI-iD behaviors are considered stable and global only when the child
also experiences comorbid aggression, then it is the comorbid aggression that needs to be
primary for intervention. However, delineating for parents the separate but co-occurring
effects of child ADHD and child aggression may be important for helping parents better
understand their children’s ADHD behaviors. Further discussion of globablity and
stability findings are included in the section pertaining to maternal role adjustment.

Another similarity in ratings across behaviors was found. Like Johnston
(Freeman, Johnston, & Barth, 1997; Johnston & Freeman, 1997), the current study found
that attributions made by mothers of children with ADHD were largely similar across
behavior type not only when examining different types of disruptive behaviors (ADHD
and disobedience) but also when examining positive/prosocial behaviors. Both in the
current study and in Johnston’s study, the patterns of attributions were similar across all

three behavior types"). Johnston specifically found that mothers of children with ADHD

'0 The current study did examine whether the ratings across behavior type were statistically similar or
different; however, the pattern of attributions was the same across behavior type.

159

did not differentiate among the behavior types when making attributions of locus and
stability but did differentiate in terms of attributions of controllability. In contrast,
parents of children without ADHD did distinguish on all behavioral dimensions
(Johnston & Freeman, 1997) and exhibited a “child-serving bias” by rating positive
behaviors as more stable and global and negative behaviors as less stable and global
(Johnston & Freeman, 1997, p. 644). Thus, they concluded that parents of children with
ADHD showed “less of a general child-serving bias” (Johnston & Freeman, 1997, p.644)
by not differentiating across behavior types and giving children more credit for positive
behaviors.

The current finding that mothers of children with ADHD attributed positive
behaviors to stable factors contrasts in part with Johnston and Freeman’s (1997)
conclusion that parents of children with ADHD attribute positive behaviors to less
dispositional factors than their counterparts. Johnston’s finding is indeed consistent with
parental negativity toward their children (in a sense giving children less “credit” for their
positive behavior). The current findings that mothers of children with ADHD rated
positive behaviors as more stable and global compared to ratings made by mothers of
children without ADI-ID contrast with this conclusion. It is possible that in the current
sample these ratings were either more adaptive for mothers (see discussion subsequently)
or that ratings were influenced by a positive “halo effect” in which ratings of one
behavior affected ratings of all other behaviors. To clarify, both the current study and
Johnston’s found both poops of mothers rated children’s positive behaviors as more
stable than unstable but which ratings were higher differed in the two studies. That is, in

the current study the stability ratings of positive behaviors were significantly higher for

160

mothers of ADHD children whereas in Johnston’s study the stability ratings were higher
for mothers of children without ADHD.

The similarity in the pattern of attributions across behavior types is somewhat
puzzling. There are several possibilities for explaining this effect. First, ratings of one
behavior may have influenced ratings of others behaviors, producing a “halo effect.”
This effect could be a “child-serving bias” if positive as well as negative behaviors were
seen as stable as found in the current study (although the similarity between ratings of
ADHD and aggression complicates this interpretation). However, in contrast to current
study findings, Johnston (Johnston & Freeman, 1997) found that parents of children
without ADI-ID did distinguish their ratings by behavior type. Further, even though
parents of children with ADHD rated positive behaviors as stable (in Johnston’s as well
as the current study), in Johnston’s work, the magnitude of those stability ratings were
greater for parents of children without ADHD. Certainly, a positive “child-serving” bias
would be beneficial for parents of children with ADHD who are faced with daily
stressors associated with their children’s behavior. As will be discussed later, the hope
in the current study was that the examination of attributions in relation to mother’s
adjustment might help elucidate the function of these different attributional patterns.
However, before directly examining attributions in relation to adjustment, one additional
concept was explored. Beliefs and expectancies are an important part of maternal
attributions and the biases they may hold for or about their children; thus, beliefs about

child diagnosis was examined in relation to attributions.

161

Maternal Attributions and Belief About Child’s Diagnosis

Because attributions are a part of a belief system, analyses examined whether
maternal attributions differed by maternal belief about child’s ADI-II) diagnosis (rather
than actual ADHD diagnostic status of the child). The results for beliefs about child
diagnosis mostly corresponded with results for actual diagnosis. That is, mothers who
believed their children to have ADHD (whether or not this was true) also attributed their
children’s disruptive behavior to more stable, global, and uncontrollable factors.
However, an interesting new finding was that the pattern of results for attributions about
positive/prosocial behaviors when using maternal 13% about diagnosis replicated
Johnston’s group differences (in contrast to the “positive halo effect”)

For the belief data, results for maternal ratings of positive behaviors differed
when separating mothers by beliefs about diagnosis versus by actual child diagnosis.
Mothers who believed their child to have ADHD were less likely to see their child’s
positive behavior as global, consistent with the “negative bias” suggested by Johnston’s
research. This finding is interesting and supports the argument that there is more of
negative than positive halo effect in the attributions of mothers of children with ADI-ID.
Further research is needed for clarification of these findings but notably, beliefs about

ADI-ID diagnosis appeared to be important to consider.

ADHD DSM-IV Subtype Differences in Attributions
One way in which the current study began to expand our knowledge about ADHD

was through the analyses of differences in maternal attributions about child misbehaviors

based on DSM—IVADHD subtype. Indeed, this was the first study to do so, in analyses

162

looking at ADHD-combined type, ADD-Inattentive type, as well as subthreshold ADHD
symptomatology. As might be expected, group effects were fiequently but not
exclusively driven by differences between ADHD-Combined subtype and controls.
Research consistently has indicated that this syndrome, which includes extensive
hyperactive symptomatology is more noticeable and disruptive to parents and others in
the child’s life than simply inattentive behaviors (Barkley, 1998; Schachar, Sandberg, &
Rutter, 1983). It is the child’s disruptive behaviors that are most obvious and that
fi'equently interfere with the plans that others in the child’s life have. Interestingly, the
role of inattentive behaviors is likely often overlooked or minimized by people in the
child’s life (Barkley, 1998). Similarly, hyperactive behaviors may be seen as more
purposefully and may be more likely to be confused with oppositional/aggressive
behaviors (Abikoff et al., 1993; Schachar, Sandberg, & Rutter, 1983).

Attributions made by mothers of children with ADD-inattentive subtype did not
differ from attributions made by mothers of children without ADHD but did differ fi'om
those made by mothers of children with ADHD-C. Thus, in general, mothers of children
with ADD-Inattentive subtype did not endorse an abnormal or unique attributional style
compared to mothers of control children. Thus, when a child has ADD-Inattentive
subtype, mothers’ attributions are very much like those attributions made when
considering a non-disordered child. Again, this data suggests that the child more overtly
disruptive behaviors (hyperactive and oppositional) may be more noticeable to mothers
and thought about differently. Notably, analyses did not examine whether attributions
difi‘ered between attributions about Inattentive versus attributions about Overactive

behaviors; an area where future research might clarify further the role of attributions in

163

relation to the various behaviors associated with ADHD, most particularly distinguishing

between inattentive and hyperactive symptomatology

Maternal Attributions and Role Adjustment

A major objective of the current study was to examine maternal attributions in
relation to maternal adjustment. This is important with regard to models of family
process in the maintenance of ADHD problems over time as well as to potential insights
into interventions aimed at ameliorating parent stress. No prior known studies with
mothers of children with ADHD focused directly on maternal attributions about
disruptive behaviors in relation to maternal adjustment. One study examined attributions
about ADHD behaviors and adjustment but was in a convenience sample through a mail
survey (Geller & Johnston, 1995). Another study (Hoza et al., 2000) measured
attributions and also adjustment factors but not as a primary focus of the study. Thus,
this was one of the first major studies of maternal adjustment in relation to maternal
attributions.

A large body of literature suggests that beliefs and attributions are related to
adjustment (Lazarus & Folkman, 1984; Podolski & Nigg, 2001; Thompson, et a1, 1992).
Thus, it was not surprising to find that maternal attributions about children’s behaviors
were indeed related to mothers’ role stress. Of more interest was the specific pattern of
these attributions. Based on the general parent attribution literature (Dix & Grusec,
1985), it was hypothesized that maternal attributions of intemality, stability, globality,
and controllability about child disruptive behaviors would be related to high levels of

maternal distress (and low levels of role satisfaction).

164

The first finding with regard to that pattern was that stability and globality were
the most consistent predictors of adjustment (when using teacher as well as maternal
ratings of child behaviors). This finding echoes the literature; that is, stability and
globality factors are the most consistent factors to be related to adjustment both in the
general literature (Bunce & Peterson, 1997) and in the parent attribution literature (Dix &
Grusec, 1985; Johnston & Freeman, 1997). Within the learned helplessness literature,
Peterson has recently argued for a two-dimensional model of “pessimistic” attributions
with the global/stable factor accounting for the most variance in relation to well-being
(Bunce & Peterson, 1997). Certainly, the stability and globality factors have been related
to poor adjustment in studies with general population samples (Bunce & Peterson, 1997)
and parents (Dix & Grusec, 1985).

The emergence of stability and globablity as robust attributions related to
maternal adjustment suggests that it is the perceived pervasiveness of child disruptive
behaviors that wears on parents rather than the factors related to whether or not the child
could change the behavior (locus and control). This finding makes sense within a
longitudinal or family development framework. Studies of coping in parents have found
that over time there is a build up of stressors such that parents need to adopt additional
means of coping (McCubbin & Patterson, 1983; Thompson & Gustafson, 1996). Given
the chronic nature of ADHD and that medication only has short-term benefits,
understanding factors which may build up over time may prove essential to effective
interventions for children and their parents. In addition to the stability of the child’s
behavior, the globality or pervasiveness of that behavior means that parents do not

receive reprieve fiom dealing with their children’s disruptive behaviors. It may be that

165

the use of medication may lessen maternal stress and thus have additional indirect
consequences for child adjustment as well.

The relation between attributions of stability and adjustment was particularly
robust for attributions pertaining to oppositional/disobedient behaviors. That is, maternal
attributions about oppositional/disobedient behaviors were related to maternal adjustment
even when controlling for child behaviors. The unique contribution of maternal
attributions to maternal adjustment supports the importance of maternal attributions in
understand adjustment to a child’s ADI-II). As outlined earlier, maternal stress plays a
significant role in determining parenting behavior (Patterson, 1983). Mothers who
experience higher levels of role specific or general stress are less likely to effectively
monitor and discipline their children. Consequential poor and coercive parenting ensues
leading to the development of aggressive behavior. Notably, in the current study, the
independent relation to maternal adjustment occurred only when considering attributions
about oppositional/disobedient behaviors not when considering attributions about
inattentive/overactive behaviors. In contrast, the relation between attributions about
inattentive/overactive behaviors and maternal role specific stress was fully mediated by
child behaviors. This indicates that attributions pertaining to oppositional behaviors may
be of utmost importance to target in interventions. Despite this pattern of results, further
study of the role of attributions about inattentive/overactive behaviors is warranted,
particularly, when considering the high correlation between attributions about
inattentive/overactive and attributions about oppositional/disobedient behaviors.

In contrast to the general robustness of the stability finding, locus and

controllability were not significantly related to maternal adjustment in the current study.

166

The roles of locus and controllability have been less clear across prior studies as well.
Yet attributions of controllability were hypothesized to be one of the crucial
differentiating factors, distinguishing attributions made by mothers of children with and
without ADI-ID. Within the developmental literature on parent attributions, Dix and
Grusec (1985) found that controllable and internal attributions of behaviors are related to
greater stress; however, this differs fi'om what has been found as the attributional pattern
made by mothers of children with ADHD (Johnston & Freeman, 1997). Given that
mothers of children with ADHD experience greater levels of distress (Johnston & Mash,
2001; Podolski & Nigg, 2001) compared to mothers of children without disorders, the
attribution pattern was expected to be moderated by diagnosis. That is, the relation
between attributions and distress was expected to be different for mothers of children

with and without ADI-ID.

Moderation of Attribution-Adj ustment Relation by Diagnosis

One of the primary questions in the current study was whether the relation
between maternal attributions about child disruptive behaviors and maternal adjustment
would differ based on child diagnosis, particularly for attributions about controllability
and locus pertaining to child ADHD and oppositional behaviors. Notably, there was
some support for moderation by diagnosis when examining attributions of stability.
Attributions that 10 behaviors were stable were associated with higher role stress for
mothers of children with ADHD but not for mothers of children without ADHD. In
contrast, maternal role satisfaction was related to attributions about child ADHD and

aggression for mothers of children without ADI-ID but not for mothers of children with

167

ADHD. The stability of ADHD behaviors was thus associated with a higher level of
increased stress for mothers of children with ADHD than for mothers of control children.
Mothers of children with ADHD are faced continuously with their children’s disruptive
behaviors that prove stressful each day. Mothers of children who do not have ADHD do
not experience the same intensity, severity, or stability of inattentive/overactive
behaviors; by definition, their children exhibit those behaviors less often and to a less
severe degree. Thus, the findings in the current study are consistent with the literature on
ADHD. Interestingly, the relation between attributions and role satisfaction was clearer
for mothers of children without ADHD. It may be that these mothers do not expect such
behaviors from their children and thus their presence is more distressing as such
behaviors are to be less expected or explicable.

Globality findings were also of interest. Globality attributions pertaining to IO
behaviors were related to lower levels of satisfaction with parenting performance,
indicating the pervasiveness of the child’s behaviors is problematic for parents. This
finding corresponds to prior research on parent stress and coping with their child’s
° ADHD (Cunningham, Bemness, & Siegel, 1988; Podolski & Nigg, 2001). Studies on
social support have found that social support was negatively associated with parent
adjustment in parents of children with ADHD. Cunningham, Bemness, and Seigel (1988)
explained this in terms of the added stress that support systems can add when they
disapprove or judge a parent.

Findings pertaining to attributions of stability were more complicated;
specifically, the relation between stability attributions and adj ustment was dependent on

child diagnosis, type of behavior, and type of maternal adjustment under consideration.

168

Notably, while both mothers of children with and without ADHD reported high levels of
m when attributing IO behaviors to stable causes, the magnitude of this relation was
constant for mothers of children with ADHD but increased for mothers of children
without ADHD the more they attributed behaviors to stable causes. Mothers of children
with ADHD may be at high enough level of stress that a type of ceiling effect is
experienced. Alternatively, the diagnostic label of ADHD and the attributions associated
with such a diagnosis may at times serve a protective function in terms of high levels of
stress further increasing. Mothers of children without ADHD may exhibit a greater
increase in role stress the more stable they see their child’s IO behaviors partly because
they are not expecting their children to act in Impulsive/Overactive ways. When
considering IO stability attributions and role satisfaction, the relation was greater (larger
magnitude and more significant) for mothers of children without ADHD. ADHD
diagnosis may serve a protective function against role specific satisfaction as mothers
understand the cause of their child’s ADHD behavior as less related to their own
parenting efforts.

Results pertaining to attributions regarding the stability of children’s
Oppositional/Disobedient behaviors showed a similar pattern. Attributions of stability of
Oppositional behaviors were related to dissatisfaction with parenting performance in
mothers of children without a behavior disorder but related to role stress in mothers of
children with a behavior disorder. Thus, although attributions of stability and globality
about 10 and OPP behaviors are related to maternal adjustment, whether they are related
to role satisfaction or role stress appears driven by child diagnosis. When children are in

the pathological range, these attributions may protect against role dissatisfaction to which

169

all parents are vulnerable; however, these attributions do not appear to protect against the
role stress generally found to be higher for parents of children with disorders.

Although locus and control attributions were expected to be related to maternal
adjustment, the current study did not find such relations. The lack of findings was not
surprising given possible confounds with medication status and given measurement
limitations. Measure limitations warrant additional investigation before conclusions are
made.

Diagnostic versus dimensional considerations

Attention Deficit Hyperactivity Disorder is a diagnosis based on number of
symptoms reaching a threshold of clinical severity. Notably, while some psychologists
argue for a clearly defined syndrome based on clinical cutoffs (Searight, Nahlik, &
Campbell, 1995), sophisticated studies indicate that whether or not ADHD is in fact a
categorical rather than dimensional phenomenon remains unclear (see Ni gg & Goldsmith,
1998), with some evidence suggesting it may represent an extreme on a continuum of
normal behaviors. The current study supports the importance of considering both
dimensional and categorical factors associated with ADHD and familial adjustment to
child attention problems. Dimensional analyses revealed that subthreshold ADHD
symptoms may prove stressful for mothers. Further, in examining the relation between
attributions and adjustment, severity of child behaviors mediated the relation between
maternal attributions about IO but not ODD behaviors. In this way, dimensional
consideration of child ADI-ID behaviors is important for a complete understanding of
contextual family factors. Diagnosis of ADHD was found to moderate the relation

between attributions about ODD behaviors and maternal adjustment, thus indicating that

170

diagnostic considerations are also important to consider in terms of maternal adjustment
as well. Thus, the current study results support the comprehensive approach of
examining child problems fiom both continuous, dimensional as well as categorical,

diagnostic perspectives (Jensen, Koretz, Locke, et al., 1993).

ADI-ID Subtype Differences in Attribution - Adjustment Relation

As with the attribution by group analyses, subtype analyses of the attribution-
adjustrnent relation indicated that results were often driven by differences between
mothers of children with ADHD-C and controls. Relations between attributions and
adjustment were not significant for mothers of children with ADD-inattentive subtype.
Interestingly, subtype analyses revealed that for mothers of children with ADHD-C
subtype, attributions of high internal locus for positive traits was related to parent
satisfaction and less stress; high internal locus for oppositional behaviors was related to
parent dissatisfaction and greater stress. Thus, attributions of locus did appear related to
adjustment but only for mothers of children with ADHD-Combined type and only when

considering Oppositional/Disobedient or positive behaviors.

Summary of Maternal Attribution Findings
In summary, the current study replicated prior research that mothers of children
with ADHD attribute their children’s disruptive behaviors to more stable, more global,
and less controllable causes. The study furthered the literature by examining subtype
differences and finding that indeed subtypes of ADHD are important when understanding

the role of attributions, especially in relation to maternal adjustment. Attributions of

171

stability and globality were consistently related to maternal adjustment whereas
attributions pertaining to controllability and locus were not. Notably, the relation
between maternal attributions and adjustment differed for diagnostic group. While
attributions of stability and globality of behaviors were related to adjustment in both
groups of mothers, these attributions were related to less role specific satisfaction in
mothers of control children and with greater m in mothers of children with ADHD.
This suggests that stability and globality attributions about 10 behaviors, attributions that
are consistent with an ADHD diagnosis, may provide some relief to mothers of children
with ADHD in terms of role specific stress and self-blame but that these attributions do

not buffer mothers fiom the general stress caused by their child’s ADI-ID behaviors.

Possible Implications Pertaining to Intervention Research

In addition to simply understanding the role of maternal attributions in relation to
maternal role adjustment, one aim of the current research is to contribute to the
effectiveness of interventions. Research has shown that how parents think about their
children’s behavior affects treatment decisions and adherence to treatments (Wright et al.,
2000). Because few studies have examined parent attributions about ADI-ID, the current
study findings were an initial foray into this arena. Similarly, few studies have examined
parent attributions about ADHD in relation to treatment decisions; however, the initial
studies that examine such notions suggest that parent attributions may play an important
role in treatment decisions and adherence (Reimers et al., 1995). Reirners and colleagues
(1995) examined attributions related to physical versus environmental causes of

children’s behavioral problems, finding that attributions to physical causes were

172

associated with a decrease in parental perception of acceptability of behavioral
treatments. Reimer’s study would suggest that attributions that correspond to a
biogenetic model of ADHD would be associated with decreased parental interest in
behavioral treatments. The current study cannot address the relation between parent
attributions and treatment decisions; however, findings suggest that ADHD behaviors are
viewed as stable and global and hence consistent with a biogenetic model of ADHD. The
findings pertaining to locus and control are inconclusive thus further research studying
these dimensions is warranted.

The primary implications of the current study are the findings pertaining to
maternal attributions of stability and globality in relation to maternal role adjustment in
mothers of children with ADHD. These two factors were the most consistent in
distinguishing mothers of children with and without ADHD and in understanding role
adjustment. These findings suggest that interventions that provide mothers with relief
from dealing with the realities of their children’s chronic and pervasive behavioral
disorder is warranted. Additionally, interventions that help mothers identify and
appreciate the breaks in their children’s disruptive and negative behaviors may be
essential with these parents. While the realities of the child’s disruptive behaviors should
not be ignored, it is likely that mothers cannot fully enjoy their child’s positive behaviors
in the midst of negative behaviors. Additionally, overactive or disruptive behaviors that
normally are not distressing to parents may be distressing to parents of children with
ADHD as their tolerance threshold for their child’s behavior may be constantly taxed.
Thus, interventions aimed at increasing positive interactions and at helping mothers

identify and enjoy such positive interactions may be important aspects of interventions.

173

Notably, many behavioral treatments for ADHD already include interventions aimed at

helping parents identify and reinforce children’s positive behaviors (e.g., Barkley, 1998).

Child Attributions

Interesting, when examining child data, attributions about locus and
controllability differed for children with and without ADHD whereas attributions about
stability and globality did not. Consistent with their disorder and with the literature,
children with ADHD rated their behaviors as more internal and less controllable
compared to children with ADHD. The difference between the patterns of results found
with mother versus child data suggests that children may focus on different aspects of
their disorder compared to their mothers. Children may be more attuned to locus
compared to their mothers. Certainly, research on internalizing disorders such as anxiety
and depression has found that children are a better source for reporting internalizing
symptoms compared to their parents (Bell-Dolan et al., 1990). Their attunement to the
internal and uncontrollable aspects of their disorder suggests that children with ADHD
think about their disorder in a way that is consistent with the diagnosis but also that is
inconsistent with Milich’s (1994) “self-protective” theory.

The result that children attributed IO behaviors to more internal causes differed
fiom Milich’s (1994) findings that children made more external attributions for “task
failures.” Of course, a primary purpose of the study (and different fi'om Milich’s work)
was to examine children attributions about their disruptive behaviors (rather than about
academic type performance). Although the current study did not test the “buffering”

hypothesis, the pattern of attributions found in the current study was inconsistent with

174

Milich’s finding that children with ADHD make external attributions for failures. In the
current study, children perceived their disruptive behaviors as internally caused. Further
research is needed in order to examine whether or not those accurate (i.e., consistent with
current diagnostic understanding of ADHD) are related to poor adjustment (i.e.,
depression, anxiety, low self-evaluation/self-esteem).

An additional contribution of the current study was the adaptation of Johnston’s
Written Analog Questionnaire for use with children. The adaptation of this measure and
examination of the pattern of attributions made by children about their own disruptive
behaviors is an initial step into understanding cognitive factors that may be related to
child adjustment. Notably, a next step for the research is to examine children’s
attributions about their own disruptive behaviors in relation to measures of adjustment
such as co-morbid anxiety and depression and measures of self-esteem.

Future directions include examination of attributions in additional social contexts.
Children with ADHD have difficulties in peer relationships (Barkley, 1998). The
examination of attributions and the effects of these attributions within interpersonal
contexts will be an important foray into understanding factors related to social
adjustment. Further studies will further the measurement of children’s attributions and

build on current study findings.

Limitations of Current Studies and Cautions when Interpreting Findings
Although the study was able to provide some new information about the role of
attributions in relation to child disruptive behaviors and maternal adjustment, findings

should be interpreted with caution due to a number of limitations. The findings that

175

replicate prior research can be considered more robust. A limitation of the current study
is a low sample size, particularly in relation to group comparisons and most particularly
for the ADD-Inattentive subtype (where sample size = 12), so that null findings could
represent type II errors.

A limitation of the current study is that mothers’ ratings of child behaviors and
mothers’ own self-reported attributions and stress were used for analyses, creating a
source-variance confound for those effects. Notably, this limitation is mitigated
somewhat by the finding that maternal adjustment was related to teacher ratings of child
behaviors as well as maternal ratings of child behaviors.

Third, findings, particularly null findings, must be interpreted with caution given
the limited reliability of some attribution scales. Reliability was especially low for the
measures of locus attributions perhaps leading to the null findings observed. The Written
Analogue Questionnaire yielded a cleaner factor solution and had better reliability than
the Recalled Incident Interview and thus was chosen as the primary measure in the study.
Despite this, the reliabilities of the locus measures were still poor (alpha=.66 for mother
WAQ ratings of IO and Oppositional behaviors; alpha=.33 and .42 for RII ratings of IO
and Oppositional behaviors). Thus, the lack of findings in the current study for mother
attributions of locus may be due at least in part to low reliability of the attribution in the
current study. Further work on these measures is in order. Additionally, although
controlling for medication status, the issues pertaining to attributions and ADHD as they
may be affected by prior knowledge about ADHD, prior knowledge about child’s ADHD

status, and the effects of attributions about medication warrant firrther consideration in

176

order to understand how those factors might affect results, particularly those pertaining to
locus but also in relation to other attribution domains.

Another possible measurement issue is that maternal attribution ratings of
prosocial/positive behaviors were in the same direction as maternal ratings of child
misbehaviors. This pattern of attributions was contrary to expectations. These findings
are especially perplexing when considering relations to maternal adjustrrrent. For
example, maternal stability and globality attributions for child positive/prosocial
behaviors were related to maternal dissatisfaction and stress. That the relations for
attributions about positive behaviors were similar to those for misbehaviors suggests

mothers may not have distinguished attributions by behavior type.

Conclusion

Despite these issues, the current study adds to the existing literature on
attributions in a number of ways. First, the current study findings replicated results
pertaining to the different pattern of attributions made by mothers of children with
ADHD compared to attributions made by mothers of children without ADHD.
Additionally, maternal attributions of stability and globality about child disruptive
behaviors (both IO and OPP) were found to be related to maternal adjustment in the
current study in directions predicted by both the attribution and ADHD literature and by
the general child literature. Significant relations between attributions of locus and
controllability were not replicated in the current study; although it is noted that the lack

of findings in the current study may be due to measurement limitations.

177

An interesting additional finding was that child diagnosis appeared to moderate
the relation between maternal attributions of stability and globality in relation to maternal
adjustment. Adding to the potential significance of this finding is that the relations
differed by child diagnostic group for the two adjustment outcome measures.
Attributions of stability and globality were related to increased global stress but not role
dissatisfaction in mothers of children with ADHD.

While these findings require replication, they suggest that attributions about child
behaviors are related to maternal role adjustment and that the ways in which they are
related differ in important ways for mothers of children with ADHD compared to
mothers of children without ADHD. The results also suggest that maternal attributions
about ADHD are commensurate with the realities of having a child with or without
ADHD. Adj ustment is related to attributions but moderated in part by child diagnosis.
Notably, attributions of stability and globality proved most predictive of maternal
adjustment. These findings echo the literature; commensurate with the realities of
parenting a child with ADHD, respite and self-care for parents of children with ADHD
may be an important factor to help parents cope with the challenges of parenting a child
with ADI-ED.

Continued work in this area may be useful in devising interventions for parents.
As mentioned above, respite and self-care activities for parents may be an important
adjunct to current treatments for child disruptive behaviors. Also, when behavior
programs are effective in reducing child disruptive behavior, helping mothers identify
and focus on those reductions may be an important cognitive/attribution addition to

primarily behavioral programs. In addition to facilitating interventions, the current study

178

elucidates further a possible mechanisms related to child outcome. An additional study
will examine maternal attributions and stress in relation to mothers’ parenting behaviors.
Additionally, future studies could examine child, peer, and teacher attributions in terms
of expectations of children, supports provided, and consequential adjustment.

Notably, while 99m children and mothers attributions differed by diagnostic
grouping of child, the attribution domains that differed were not the same for mothers and
children. Children rated their own IO behaviors as internal and uncontrollable.
Additional tests for future studies would directly compare mother and child attributions to
examine if attributions by behavior type were significantly different within parent-child
dyad. This would allow us to understand how attributions may or may not be passed
down from mother to child. However, the current study findings suggest that children
and mother think about child behaviors in different ways (or at least that they focus on
different aspects of the behaviors). No conclusions can be made about a possible
“protective” or “deleterious” relation between children’s attributions as child attributions
were not tested in relation to adjustment. A future study will be to examine children’s
attributions about their disruptive behaviors in relation to co-morbid internalizing
symptoms (i.e., anxiety and depression); that study will provide more data relevant to the

implications of the different attributions made by children with and without ADHD.

179

APPENDICES

180

Appendix A

 

 

 

 

 

 

Table 65 Studies pertaining to attributions made b I children
Study Sample Attributions Found to be Attributions Found to be Event studied
for negative related to for positive related to
events events
Peterson Normal Pessimistic Quitting Optimistic Optimism Uncontrollabl
(Peterson et adults explanatory behaviors explanatory Good e events;
al., 1998) style = Depression style = physical Measured
stable, Poor physical Stable, health attributions
internal, health internal, about positive
global, specific and negative
uncontrollabil attributions events
ity for positive separately
attributions events
for negative
events
Peterson & College Students who Attributions
Barrett students made about
(1987) external, bad/negative
unstable, and academic
specific events
attributions
for bad
events
received
higher grades
May be part
of optimistic
style
Sweeney, Meta-analytic Attributions Attributions
Anderson, & review, to internal, to external,
Bailey (1986) studies with stable, and unstable, and
nonnal adults global were specific for
related to positive
depression events were
related to
depression
Ability and
luck
attributions
for positive
were related
to depression
Dweck Non- Low internal Lack of High internal More effort Internal
(Dweck & disordered responsibility effort; responsibility responsibility
Leggett, children ; increased & for academic
1988) High external quitting Low external success and
responsibility responsibility failures
attributions for combined combined
for combined negative and
negative and positive Note: did not
positive events differentiate
events positive and
Tanned negative
Termed - events
“helpless” oriented”
Mischel, Internal Ineffective Academic
Zeiss, & expectancies activity type activities

 

 

 

 

 

 

 

 

181

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Zeiss (l 974) for negative
events
Licht et al. Children with Attributions Poor Persistence
(1985) learning to insufiicient perfonmnce on reading
disabilities ability on task
(internal) persistence
readingztask
Carlson et al. Children with High internal Poor “Negative” Academic
(2000) and without attributions performance non- task
ADHD protective
attributions
were made by
ADHD
children for
positive
events as
exhibited by
low internal
& high
external
attributions,
and relation
to poor
perforrmnce
Milich Boys with External Increased Puzzle
(Milich & ADHD attributions effort solving tasks
Okazaki, for failure
I991)
Hoza, 83 boys with Boys with Boys with Puzzle
Pelham, ADHD ADI-ID made ADHD made solving task
Waschbusch, 66 non- less internal more external
Kipp, & disordered (more attributions
Owens, boys external) for for successes
(2001) failures
Hoza, 27 boys with Less likely to . Boys with Study of self-
Pelham, ADHD attribute ADHD perceptions
Milich, 25 non- negative attributed and
Pillow, & disordered social positive attnhutions in
McBride boys outcorrres to social relation to
(1993) internal outcomes for social
factors internal outcomes.
Low sarrple
size
Johnston & Non- Noncomplian ADI-ID more Attributions
Leung (2001) disordered ce more controllable about
children intentional when compliance
viewing tape than ADHD behavioral & when
of ADHD behaviors meds; treatmart
behaviors involved
medication,
medication
and
behavioral,
no treatment
Johnston, Children with Compliance Attributions
Fine, Weiss, ADI-ID attributed to about
Weiss, Weiss, ability, effort, behavior
& Freeman and task when
(2000) when off- medicated
medication; and not
medicated for
ADHD

 

182

 

Appendix B

Table 66 Studies pertaining to attributions made b

 

 

 

 

 

 

 

 

 

 

 

 

 

 

I parents
Study Sample Attrrhutional Relation to Attributions] Relation to Attributions
style related emotional or style related emotional or pertained to
to negative behavioral to positive behavioral
events resmnse events response
Weiner College Internal and Low help-
(1980) students controllable giving
attributions
Attributions
that need was
external and
uncontrollabl
e were
associated
with positive
emotional
response
Dagnan, Care staff When patient Were Negative
Trower, & workers of negative associated patient
Smith (1998) severity behaviors with negative behaviors
disabled rated as emotion and
adults controllable less
willingness to
help
Dix & Grusec Parents of Child Such Child
(1985) non- misbehaviors attributions nrisbehaviors
disordered rated as for negative
children Internal, child
controllable, behavior
and stable were rated as
likely to be
“Pseflifi
Geller & Mothers of Child When child Child .
Johnston nonselected nonconrplian noncomplian noncomplian
(1997) community ce ce was rated ce;
children When rated as more In survey
as more internal and study,
internal and controllable examined
controllable were related attributions in
were related to less relation to
to less investment in self-reported
investment in parenting expected
parenting emotional
and
behavioral
response
Gretarsson & Mothers of Child Positive child Child
Gelfand non- misbehaviors behaviors behaviors
(1988) disordered were rated as were rated as
children external, internal,
uncontrollabl controllable,
e, and and stable
transient
Baden & Mothers of Misbehaviors Child
Howe (I992) boys with seen as due to nrisbehaviors
conduct stable and
disorders global causes
Rated as not
controllable

 

183

 

 

by parent

 

 

 

 

 

 

 

 

Johnston and Parents of Hyperactive Child
Freeman children with and misbehaviors;
(I997 ADHD oppositional ratings by
behaviors parents of
seen as less children with
controllable ADI-ID were
and more compared to
stable ratings by
parents of
controls
Johnston et Mothers of Negative Compared
al. (2000) children with behaviors ratings of
ADHD were rated as mothers when
more child on and
externally ofi'
caused when medication
child
medicated
Hoza et al. Parents of Child Were related Child
(2000) children with misbehavior to negative
ADI-ID rated as due Low behaviors
to low child parenting
effort eflicacy and
dysfunctional
discipline

 

 

 

184

 

Appendix C

Table 67 Test of differences in attributions made by mothers of children with ADHD
diagnosis versus mothers of children without a behavioral disorder, using the R11 measure

that describes child behavior problems.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Control ADHD any F statistic (df) P value Value on Attributions
type attrrh about ADHD
measure group in
comparison
to control
RII IO 4.7 6.0 4.9 P<.05 l=unstable ADHD more
stability (2.4) (2.2) (1,65) lO-stable stable
RH 10 5.2 7.9 22.8 P<.001 1=specific ADI-ID more
_globality (2.5) (2.0) (L65) 10=global gobs]
R11 10 locus 6.9 7.5 1.1 n.s. 1=external
(2.8) (2.4) (1,65) 10=intemal
R11 10 7.9 6.0 10.8 P<.01 l=uncontrolla ADHD less
control (2.5) (2.2) (1,65) ble controllable
lO-controllab
le
RII OPP 3.8 5.2 5.6 P<.05 l=unstable ADHD more
stabiliy (2.6) (2.2) (1,76) lO-stable stable
R11 OPP 5.5 6.5 2.3 n.s. laxtemal ADHD more
locus (2.8) (2.5) (1,67) 10=internal internal
RII OPP 8.9 7.5 15.2 P<.001 1=uncontrolla ADI-ID less
control (1.2) (1.7) (1,67) ble controllable
10=controllab
1e
RII Positive 9.7 9.4 2.3 n.s. l=uncontrolla
control (.75) (.89) ( 1,67) ble
10=controllab
le

 

*IO='Inattentive-Overactive type of behavior
‘OPP=oppositional/disobedient type of behavior

185

 

D
Table 68 Results of maternal attributions about child behavior, testing for group

differences when comparing the independent variable ofghild diagnosis with four levels
(ADHD-C, ADD, subthreshold, and controls), dependent variable=attribution domain by

 

 

 

 

 

 

 

 

 

 

 

behavior e. Using Recalled Incident Interview to measure attributions
a b C
Control ADI-ID- ADD - Sub- F stat p-value Value on Conclu
Combine Inattentiv threshold (df) attrib -
d e ADHD measure Slons
R11 10 4.7 6.0 5.0 5.4 1.7 n.s. l-unstabl
Stability (2.4) (2.2) (2.8) (2.3) (3,1 l 1) e
__ 10—stab1e
R11 10 5.21'c 7.91' 7.1 7.6c 11.1 P<.001 1=specifi Control
globality (2.5) (2.0) (2.0) (1.6) (3,1 1 1) c and
lO-global ADHD
difl'er
Control
and
subtlrresh
old difl‘er
R11 10 6.9 7.5 7.9 7.5 .68 n.s. 1=externa
locus (2.8) (2.4) (2.9) (2.5) (3,1 10) l
lO-intern
al
R11 10 7.9 " 6.0' 7.3 6.7 3.4 P<.05 l=uncontr Control
control (2.5) . (2.2) (2.7) (2.7) (3,111) ollable and
10=contro ADHD—C
llable
RII OPP 3.8 5.2 5.2 5.0 2.5 P<.1 1=unstab1
stability (2.6) (2.2) (2.9) (2.2) (3,1 13) e
10=stable
RII OPP 5.5 6.5 6.3 . 6.8 1.5 n.s. l=extema
locus (2.8) (2.7) (2.9) (2.5) (3,1 13) l
lO-intern
a1
RII OPP 8.9 ' 7.5 ' 8.8 8.6 5.7 P<.001 l-uueontr Control
control (1.2) (1.7) (1.9) (1.3) (3,113) ollable and
10=contro ADHD-C
llable
RII 9.7 9.4 9.8 9.6 1.2 n.s. l=uncontr
Positive (.75) (.79) (.58) (.84) (3,1 13) ollable
control 10=contro
llable

 

 

 

 

 

 

 

 

 

 

‘IO=Inattentive-Overactive type of behavior

'OPP=oppositional/disobedient type of behavior
it indicates significant difference between control and adhd combined groups
b indicates significant difference between control and add inattentive groups
c indicates significant difference between control and “subthreshold” groups
(1 indicates significant difference between adhd and add group
e indicates significant difi'erences between add and subthreshold groups
1' indicate significant difference between adhd and “subthreshold groups

186

 

Appendix B

Table 69 Correlations between maternal attributions (using scenario measure, RH) and

 

 

 

 

 

 

 

 

 

 

 

 

 

 

r arent role stress and satisfaction
Mothers Perception of Mothers PSI Total Stress
Parenting Performance
R11 10 stability -.12 .21"I
RH 10 globality -.19* 37*"
R11 IO locus .07 -.06
R11 10 control -.04 -.09
R11 OPP stability -.16+ .26"
R11 OPP locus -.17 .21“
RH OPP control -.03 -.l9*
RII Positive control .04 -.10

 

mp<.oor, ”p<.01, ‘p<.05

187

 

Appendix F

Table 70 Results with R11. Correlations between maternal attributions about child
behavior in relation to mother and teacher ratings of child behaviors (3 -factor analysis)

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Mom rating Mom rating Mom rating Teacher Teacher Teacher
of child of child of child rating of rating of rating of child
Inattention hyperactivity OPP/disobedi child child OPP/disobedi
ent Inattention Hyperactivity ent
Behavior Behavior

R11 10 .19" .22“ .16+ -.00 .07 .03

stability

RII IO .47‘” .42*" .46‘" .23“ .25" .18+

lobality '

R11 10 locus .14 .03 -.00 0 -.09 -. 17+

R11 10 -.20‘I -.14 .00 -.14 -.22"' -.04

control

RII ODD .22“ .26“ .20“ .06 .12 .09

stability

R11 ODD .26" .17+ .20“ .16 .15 .12

locus

R11 ODD -.21‘ -.22“ -.06 -.06 -.22"' -.02

control

RII Positive -.06 -.05 .01 -.14 -.17+ .01

control

"‘p<.001, "p<.01, *p<.05, +p<.l

188

 

Appendix G

Table 71 Regression to test mediation of child behavior and maternal role adjustment by
maternal attributions with dependent variable=PSI-SF Total Stress, controlling only for
child behavior corresponding to attribution)

 

Predictor
variable

Betastepl

Betastepl

Beta step3

R1 Change
Step 1

R2 Change
Step2

R2 Change
Step 3

 

Model 1

Child
Inattn

.19"

.14

.16

.44...

 

Child Hyp

‘52...

.49...

.55...

 

Globality
about 10
behavior

.15+

.18+

.02+

 

Interaction
term

-.10

 

 

Model 2

Child
Inattn

.19+

.16

.16

.44...

 

Child Hyp

.52...

.50‘..

.51...

 

Stability
about 10
behavior

.10

.10

.01

 

Interaction
term

 

 

Child
Disobedien
t/
Opposition
al

.67m

.58...

.57...

.45m

 

Stability
about ODD
behavior

.18‘

.16

.03‘

 

Interaction
term

.00

 

 

 

 

 

 

 

 

 

 

mp<.oor, ”p<.01, ‘p<.05, +p<.l

189

 

REFERENCES

Abidin, RR. (1995). Parenting Stress Index (3rd ed). Odessa: Psychological
Assessment Resources, Inc.

Abikoff, H. (1991). Interaction of methylphenidate and multimodal therapy in
the treatment of attention deficit hyperactivity disorder. In B. Osman & L.L. Greenhill
(Eds), Rit_alin: Theory and patient management. New York: My Ann Liebert.

Abikoff, H., Courtney, M., Pelham, W.E., & Koplewicz, HS. (1993). Teachers’
ratings of disruptive behaviors: The influence of halo effects. Journal of Abnormal

Child Psychology, 21, 519-533.

Abikoff, H., & Gittelman, R. (1985). Hyperactive children treated with
stimulants: Is cognitive training a useful adjunct? Archives of General Psychia_try, 42,
953-961.

Abramson, L.Y., Seligrnan, M.E.P., & Teasdale, J .D. (1978). Learned
helplessness in humans: Critique and reformulation. Journal of Abnormal Psychology,
8_7_, 49-74.

Achenbach, TM. (1991). Manual for the Child Behavior Checklist/4-18 and 1991
Profile. Burlington, VT: University of Vermont, Department of Psychiatry.

Allen, J ., & Drabman, R. (1991). Attributions of children with learning
disabilities who are treated with psychostimulants. Learning Disabilig Quarterly, 14, 75-
79.

American Psychiatric Association. (1994). Diagnostic and Statistical Manual of
Mental Disorders(4th Ed). Washington, DC: American Psychiatric Association Press.

Anderson, J .C., Williams, S., McGee, R., & Silva, A. (1987). DSM-III disorders
in pre-adolescent children: Prevalence in a large sample from the general population.
Archives of General Psvchia‘trv. 44. 69-76.

Anastopoulos, A.D., Guevremont, D.C., Shelton, T.L., & DuPaul, G]. (1992).
Parenting stress among families of children with Attention Deficit Hyperactivity
Disorder. Journal of Abnormal Child Psychology, 20, 503-519.

Anastopoulos, A.D., Shelton, T.L., DuPaul, G.J., & Guevremont, DC. (19930.
Parent training for attention-deficit hyperactivity disorder: Its impact on parent
functioning. Journal of Abnormal Child Pjsvchology, 21, 581-596.

Baden, A.D. & Howe, G.W. (1992). Mothers’ attributions and expectancies

regarding their conduct-disordered children. Journal of Abnormal Child Psychology, 20,
467-485.

190

Baker, DB. (1994). Parenting stress and ADHD: A comparison of mothers and
fathers. Journal of Emotional and Behavioral Disorders 2 46-50.

 

Barkley, R.A. (1998). Attention Deficit Hyperjactivitv Disorder: Handbook for
Diagrrosis and Treatment (2”I Ed). New York: Guildford Press.

Barkley, R.A. (1997). Defiant children 2'“I edition. : A clinician’s manual for
assessment and parent training. New York: Guildford Press.

Barkley, R.A. (1990). Attention-Deficit Hyperactivity Disorder: A handbook for
diaggosis and treatment. New York: Guildford Press.

Bell-Dolan, D.J., Last, C.G., & Strauss, CC. (1990). Symptoms of anxiety
disorders in normal children. J ourngl of the American Academy of led and Adolescent
Psychiag, 29, 759-765.

Biederman, J ., Faraone, S.V., Hatch, M., Mennin, D., et al., (1997). Conduct
disorder with and without mania in a referred sample of ADHD children. Journal of
Affective Disorders, 44(2-3), 177-188.

Biederman, J ., Faraone, S.V., Weber, W., Russell, R.L., Rater, M., & Park, KS.
(1997). Correspondence between DSM-III-R and DSM-IV attention-deficit/hyperactivity

disorder. Journal of the American Academy of Child and Adolescent Psychim, 36112),
1682-1687.

Barkley, R.A. & Cunningham, CE. (1979). The effects of methylphenidate on the
mother-child interactions of hyperactive children. Archives of General Psychiagg, 36,
201-208.

Befera, M.S. & Barkley, R.A. (1984). Hyperactive and normal girls and boys:
Mother-child interactions, parent psychiatric status and child psychopathology. Journal

of Child Psychology and Psychiagy, 26, 439-452.

Bickett, L. R., Milich, R. & Brown, RT. (1996). Attributional styles of
aggressive boys and their mothers. Journal of Abnormal Child Psghology, 24, 457-472.

Biederman, J ., Faraone, S.V., Keenan, K., Benjamin, J ., Krifcher, B., Moore, C.,
Sprich-Buckrninster, S., et al., (1992). Further evidence for family-genetic risk factors in
Attention Deficit Hyperactivity Disorder: Patterns of comorbidity in probands and
relatives in psychiatrically and pediatrically referred samples. Archivesfof General

ngchiat_ry, 49, 728-738.

Biederman, J ., Faraone, S., Mick, E., Moore, P., & Lelon, E., (1996). Child
Behavior Checklist findings firrther support comorbidity between ADHD and major

191

depression in a referred sample. Journal of the American Academy of Child and
Adolescent Psychiggy, 35, 734-742.

Biederman, J ., Milberger, S., Faraone, S.V., Kiely, K., Guite, J ., Mick, E., Ablon,
J .S., Warburton, R., Reed, E.,& Davis, SC. (1995). Impact of adversity on functioning
and comorbidity in children with attention-deficit hyperactivity disorder. Journal of

American Academic Child and Adolescent Psychiagy, 34, 1495-1503.

Biederman, J ., Faraone, S.V., Keenan, K., & Tsuang, M.T. (1991). Evidence of
familial association between attention deficit disorder and major affective disorder.

Archives of General Psychiagy, 48, 633-642.

Biederman, J ., Newcorn, J ., & Sprich, S. (1991). Comorbidity of attention deficit
hyperactivity disorder with conduct, depressive, anxiety, and other disorders. American

Journal of Psychiag, 148, 564-577.

Breen, M., & Barkley, R. (1984). Psychological adjustment in learning disabled,
hyperactive, and hyperactive/learning disabled children using the Personality Inventory
for Children. Journal of Clinical Child Psychology, 13, 232-236.

Brewin, CR. (1985). Depression and causal attributions: What is their relation?
Psychological Bulleg'g, 98, 298-309.

Bugental, D.B. & Johnston, C. (2000). Parental and child cognitions in the
context of the family. Annual Review of Psychology, 51, 315-344.

Bugental, D.B., Johnston, C., New, M., Silvester, J. (1998). Measuring parental
attributions: conceptual and methodological issues. Journal of Fgmilv Psychology, 12,
459-480.

Bugental, D.B., & Shennum, W.A. (1984). “Difficult” children as elicitors and
targets of adult communication patterns: An attributional-behavioral transactional

analysis. Monoggphs of the Society for Research in Child DevelopmenL 49 (1, Serial
No. 205)., pp. 1-69.

Bunce, S.C. & Peterson, C. (1997). Gender differences in personality correlates
of explanatory style. Personality Ed Individual Differences, 23(4), 639-646.

Brooks-Gunn, J. (1985). Maternal beliefs about children’s sex-typed
characteristics as they relate to maternal behavior. In LE. Sigel (Ed.). Parental Belief

Sfitems: the psychological consequences for children. (pp.3l9-344). Hillsdale, NJ:
Lawrence Erlbaum Associates.

Carlson, C.L., Mann, M., & Alexander, D.K. (2000). Effects of reward and
response cost on the performance and motivation of children with ADHD. Coggitive

Thmpy and Research, 24, 87-98.

192

Carlson, C.L., Pelham, W.E., Milich, R., & Hoza, B. (1993). ADHD boys’
performance and attributions following success and failure: Drug effects and individual

differences. Cogm'tive Thmy and Research, 17, 269-544.

Christophersen, E.R. & Mortweet, S.L. (2001). Treatments that work with

children: Empiricallv supported strateg' es for manggm' g childhood problems.
Washington, DC: American Psychological Association.

Cohen, J. (1992). A power primer. Psychological Bulleg,’ 112, 155-159.

Cohen, J. (1988). Statistical p_cwcr anemic for the behavioral sciences. (2“I
ed.). Hillsdale, NJ: Erlbaum.

Conners, CK. (1997). Conners’ Rating Scales-Revised: Instruments for use
with children and adolescents. New York: Multi-Health Systems, Inc.

Conners, CK. (1990). Conners’ Rating Scales Manual: Instruments for use with
children and adolescents. Toronto, Ontario: Multi-Health Systems, Inc.

Crandall, V.C., Katkovsky, W., & Crandall, V.J. (1965). Children’s beliefs in
their own control of reinforcements in intellectual-academic achievement situations.

Child Development, 36, 91-.109.

Crick, N.R., & Dodge, K.A. (1994). A review and reformation of social
information-processing mechanisms in children’s social adjustment. Psychological
Bulletip, 115, 74-101.

Cunningham, C.E., Bemness, B.B., & Siegel, LS. (1988). Family functioning,
time allocation, and parental depression in the families of normal and ADHD children.
Journal of Clinical Child Psychology, 17, 169-177.

Cunningham, C.E., Bremner, R., & Boyle, M. (1995). Large group community
based parenting programs for families of preschoolers at risk for disruptive behaviour
disorders: Utilization, cost effectiveness and outcome. Journal of Child thology and

Psychiafl, 36, 1141-1159.

Dagnan, D., Trower, P., & Smith, R. (1998). Care staff responses to people with
learning disabilities and challenging behavior: A cognitive-emotional analysis. British
Journal of Clinical Psychology, 37, 59-68.

Danforth, J .S., Barkley, R.A., & Stokes, TR (1991). Observations of parent-
child interactions with hyperactive children: Research and clinical implications. Clinical
Psychology Review. 1 1. 703-727.

193

Dedi, E.L., & Ryan, R.M. (1984). Intrinsic motivation and self-determination in
human behavior. New York: Plenum Press.

Dekovic, M., Gerris, J .R.M., Janssens, J .M.A.M. (1991). Parental cognitions,
parental behavior, and the child’s understanding of parent-child relationship. Merrill-

Palmer Merly, 37, 523-541.

Diener, M.B., & Milich, R. (1997). Effects of positive feedback on the social
interactions of boys with attention deficit hyperactivity disorder: A test of the self-

protective hypothesis. Journal of Clinical Child Psychology, 26, 256-265.

Dix, T.H., & Grusec, J .E. (1985). Parent attributions processes in the
socialization of children. In I. E. Siegel, Parental Belief Systems: The psyshological
consguences for children. New Jersey: Lawrence Erlbaum Associates.

Dix, T., & Lochman, J. (1990). Social cognitive and negative reactions to
children: A comparison of mothers of aggressive and nonaggressive boys. Journal of
Social and Clinical Psychology, 9, 418-438.

Dodge, K. (1980). Social cognition and children’s aggressive behavior. Child
Development. 51. 162-170.

Dodge, K.A. (1985). Attributional bias in aggressive children. In P. Kendall

(Ed.), Advances in coggu'five-bellsvioral research and therapy (Vol. 4, pp. 74-111). New
York: Academic Press.

Dodge, K.A., & Coie, J .D., (1987). Social-inforrnation-processing factors in
reactive and proactive aggression in children’s peer groups. Journal of Personalig and
Social Psychology, 53(6), 1146-1158.

Donovan, W.L., Leavitt, L.A., & Walsh, R.O. (1990). Maternal self-efficacy:
illusory control and its effect on susceptibility to learned helplessness. Child

DevelopmenL 61, 1638-1647.

Douglas, V.I. (1985). The response of ADD children to reinforcement:
Theoretical and clinical implications. In L.M. Bloomingdale (Ed.), Attention Deficit
Disorder: Identification, Course, and Treatment Rationale (pp. 49-65). New York:
Spectrum Publications.

DuPaul, G.J. & Barkley, R.A. (1990). Medication therapy. In R.A. Barkley

(Ed), Attention deficit hyperactivig disorder: A handbook for diagposis and treatment
(2“l ed., pp. 573-612). New York: Guildford Press.

DuPaul, G.J., Power, T.J., Anastopoulos, A.D., & Reid, R. (1998). ADHD
Rating Scale-IV: Check_lists. Norms and Clinical Interpretaytion. New York: Guildford
Press.

 

194

Dweck, CS. (197 5). The role of expectations and attributions in the alleviation
of learned helplessness. Journal of Pmnality and Social Psychology, 31, 674-685.

Dweck, C.S., Goetz, T.E., & Strauss, N.L. (1980). Sex differences in learned
helplessness: IV. An experimental and naturalistic study of failure generalization and its

mediators. Journal of Personalng spd Social Psychology, 38, 441-452.

Dweck, C.S., & Leggett, EL. (1988). A social-cogrritive approach to motivation
and personality. Psychological Review. 95. 256-273.

Dweck, C.S., & Reppucci, ND. (1973). Learned helplessness and reinforcement
responsibility in children. Journal of Personalipy and Social Psychology, 23(1), 109-116.

Edwards, M.C., Schulz, E.G., & Long, N. (1995). The role of the family in the
assessment of attention deficit hyperactivity disorder. Clinical Psychology Review, 15
LS), 375-394.

Faraone, S.V., Biederman, J ., Chen, W.J., Milberger, S., Warburton, R., &
Tsuang, M.T. (1995). Genetic heterogeneity in Attention-Deficit Hyperactivity
Disorder: Gender, psychiatric comorbidity, and maternal ADI-II). Journal of Abnormal
Psychology, 104, 334-345.

Faraone, S.V., Biederman, J ., Jetton, J .G., & Tsuang, M.T. (1997). Attention
deficit disorder and conduct disorder: Longitudinal evidence for a familial subtype.
Psychological Medicine. 27(2). 291-300.

Fiske, S.T., & Taylor, SE. (1991). Social Coggp'tion (2"d Edition). New York:
McGraw-Hill.

Freeman, W.S. & Johnston, C. (2001). Parents’ attributions for inattentive,
impulsive and oppositional child behaviors and the impact of behavioral context.
Presentation made at the bi-annual meeting of the International Society for Research on
Child and Adolescent Psychology. Vancouver, British Columbia.

Freeman, W.S., Johnston, C., & Barth, F. (in press 1997). Parent attributions for
inattentive-overactive, oppositional-defiant, and prosocial behaviors in children with

ADHD. Canadian Journal of Behavioral Science.

Frick, P.J. (1994). Family dysfirnction and the disruptive behavior disorders: A
review of recent empirical findings. Advances in Clinical Child Psychology, 16, 203-
225.

Frick, P.J, & Jackson, Y.K. (1993). Family functioning and childhood antisocial

behavior: Yet another reinterpretation. Journal of Clinics] Child Psychology, 22, 410-
419.

195

Frick, P.J., Lahey, B.B., Loeber, R., Stouthamer-Loeber, M., Christ, M.G., &
Hanson, K. (1992). Familial risk factors to oppositional defiant disorder and conduct
disorder: Parental psychopathology and maternal parenting. Journal of Consulting and

Clinical Psychology, 60, 49-55.

Friedling, C. & O’Leary, SC. (1979). Effects of self-instructional training on
second- and third-grade hyperactive children: A failure to replicate. Journal of Applied
Behavior Analysis, 12, 21 1-219.

Gage, J .D. & Wilson, L.J. (2000). Acceptability of attention-deficit/hyperactivity

disorder interventions: A comparison of parents. Journal of Attention Disorders, 4(3 ),
174-182.

Geller, J. & Johnston, C. (1995). Predictors of mothers’ responses to child

noncompliance: Attributions and attitudes. Jomnal of Clinical Child Psyshology, 24,
272-27 8.

Gomez, R., & Sanson, A.V. (1994). Mother-child interaction and
noncompliance in hyperactive boys with and without conduct problems. Journal of Child

Psychology and Psychiagy, 35, 477-490.

Graham, 8., Hudley, C., & Williams, E. (1992). Attributional and emotional
determinants of aggression among Afiican-American and Latino young adolescents.

Developmental Psychology, 28, 731-740.

Greenhill, L.L., Swanson, J .M., Vitiello, B., Davies, M., Clevenger, W., Arnold,
L., Abikoff, H.B., Bukstein, O.G., Conners, C.K., Elliott, G.R., Hechtrnan, L., Hinshaw,
S.P., Hoza, B., Jensen, P.S., Kraemer, H.C., March, J .S., Newcorn, J .H., Severe, J .B.,
Wells, K., Wigal, T. (2001). Impairment and deportment responses to different
Methylphenidate doses in children with ADI-ID: The MTA titration trial. Journal of
American Academy of Child and Adolescent Psychiap'y, 40, 180-187.

Gretarsson, S.J., & Gelfand, D.M. (1988). Mothers’ attributions regarding their
children’s social behavior and personality characteristics. Developmental Psychology,
25, 264-269.

Guidubaldi & Cleminshaw (1994). Parenting Satisfaction Scale. San Antonio:
The Psychological Corporation.

Hart, S.D. & Hare, RD. (1997). Psychopathy: assessment and association with
criminal conduct. In D.M. Stoff, J. Breiling, & J .D. Maser (Eds). Handbook of
Antisocial Behavior pp. 22-35, New York: John Wiley & Sons, Inc.

 

196

Harris, M.J., Milich, R., Corbitt, E.M., Hoover, D.W., et al. (1992). Self-
fulfilling effects of stigmatizing information on children’s social interactions. Journal of

Personalig and Social Pflchology, 63 (1), 41-50.

Harvey, E. (2000). Parenting similarity and children with attention-
deficit/hyperactivity disorder. Child and Family Behavior Therspy, 22, 3964

Harvey, E. (1998). Parental employment and conduct problems among children
with attention-deficit/hyperactivity disorder: An examination of childcare workload and
parenting well-being as mediating variables. J oumsl of Social and Clinical Psychology,
$476490.

Hechtrnan, L. (1981). Families of hyperactives. In R.G. Simmons (Eds),
Research in communitv mentslr heafl (pp.275-292). Greenwich, CT: JAI.

Hechtrnan, L., Weiss, G., Perlman, T., & Amsel, R. (1984). Hyperactives as
young adults: initial predictors of adult outcome. Journal of the American Academy of
Child and Adolescent Psychiamg, 23, 250-260.

Hechtrnan, L., & Weiss, G. (1983). Long-term outcome of hyperactive children.
American Journal of Orthopsychiat_ry, 53, 532-541.

Heyman, G.D., & Dweck, CS. (1998). Children’s thinking about traits:
Implications for judgments of the self and others. led Development, 69(2), 391-403.

Hinshaw, SP. (1994). Attention deficits and hyperactivity in children,
Developmentgl Clinicsl Psychology and Psychiat_ry, 29, London: Sage Publications.

Hinshaw, S., Henker, B., Whalen, C., Ehrardy, D., & Dunnington, RE. (1989).
Aggressive, prosocial and nonsocial behavior in hyperactive boys: Dose effects of MPH
in naturalistic settings. Journal of Consulting and Clinical Psychology. 57(4),_636-643.

Hinshaw, S.P., Zupan, B.A., Sirnmel, C., Nigg, J .T., & Melnick, S. (1997). Peer
status in boys with and without attention-deficit hyperactivity disorder: Predictions fiom
overt and covert antisocial behavior, social isolation, and authoritative parenting beliefs.
Child Development, 68(5), 880-896.

Hoza, B., Owens, J .S., Pelham, W.E., Swanson, J .M., Conners, C.K., Hinshaw,
S.P., Arnold, L.E., & Kraemer, BC (2000). Parent cognitions as predictors of child
treatment response in Attention-Deficit/Hyperactivity Disorder. Journal of Abnormal

led Psyphology, 28, 569-585.

Hoza, B. & Pelham, W.E. (1995). Social-cogrritive predictors of treatment
response in children with ADHD. Journpl of Social and Clinical Psychology, 14 (1), 23-
35.

 

197

Hoza, B., Pelham, W.E., Milich, R., Pillow, D., & McBride, K. (1993). The self-
perceptions and attributions of attention deficit hyperactivity disordered boys and
nonreferred boys. Journal of Abnormal Child Psychology, 21, 271-286.

Hoza, B., Pelhanr, W.E., Waschbusch, D.A., Kipp, H., & Owens, J .S. (2001).
Academic task persistence of normally achieving ADHD and control boys: Performance,

self-evaluations, and attributions. Journal of Consulting and Clinical Psychology, 69,
27 1-283.

Hudley, C. & Graham, S. (1993). An attributional intervention to reduce peer-
directed aggression among Afiican-American boys. Child Developmeng, 64, 124-135.

Huesmann, R.L., Eron, L.D., Leflrowitz, M.M., Walder, L.O. Stability of
aggression over time and generations. Developmental Psychology, 20, 1120-1134.

Ialongo, N.S., LOpez, M., Horn, W.F., Pascoe, J .M., & Greenberg, G. (1994).
Effects of psychostimulant medication on self-perceptions of competence, control, and
mood in children with attention deficit hyperactivity disorder. Journal of Clinical Child

Psychology, 23, 161-173.

Iverson, T.J. & Segal, M. (1992). Social behavior of maltreated children:
Exploring the links to parent behavior and beliefs. LE. Siegel, A.V. McGillicuddy-
DeLisi, & J .J . Goodnow (Eds), (pp. 267-289). Parental Belief Systems: the

psychological consmuences for children. (2“d Edition). Hillsdale, NJ: Lawrence

Erlbaum.

Jenson, C.E., Green, R.G., Singh, N.N., Best, A.M., & Ellis, CR. (1998).
Parental attributions of the causes of their children’s behavior. Journal of Child and
Family Studies, 7, 205-215.

Jensen, J .B., Burke, N., & Garfinkel, RD. (1988). Depression and symptoms of
attention deficit disorder with hyperactivity. J oumal of the American Academy of CE 1d

and Adolescent Psychiap'y, 27, 742-747.

Jensen, P.S., Martin, D., & Cantwell, DP. (1997). Comorbidity in ADHD:
Implications for research, practice, and DSM-V. Journal of the American Academy of
Child and Adolescent Psychiagl, 36, 1065-1079.

Jensen, P.S., Shervette, R.E. III, Xenakis, S.N., & Richters, J. (1993). Anxiety
and depressive disorders in attention deficit disorder with hyperactivity: New findings.
America_n Journal of Psychiag, 150, 1203-1209.

Johnston, C. (1996). Addressing parent cognitions in interventions with families

of disruptive children. In K.S. Dobson, K.D. Craig, (Eds), Advances in Coggp'tive-
Behavioral Thegpy, pp. 193-209. Thousand Oaks, CA: Sage.

198

Johnston, C. (1996). Parent characteristics and parent-child interactions in
families of nonproblem children and Children with ADHD with higher and lower levels
of oppositional-defiant behavior. Journal of Abnormal Child Psychology, 24, 85-104.

Johnston, C. & Freeman, W. (1997). Attributions for child behavior in parents of
children without behavior disorders and children with Attention Deficit-Hyperactivity
Disorder. Journal of Constplting and Clinical Psychology, 65, 636-645.

 

Johnston, C., Fine, S., Weiss, M., Weiss, J ., Weiss, G., & Freeman, W.S. (2000).
Effects of stimulant medication treatment on mothers’ and children’s attributions for the
behavior of children with attention deficit hyperactivity disorder. Journal of Abnormal

Child Psychology, 28, 371-382.

Johnston, C. & Leung, D.W. (2001). Effects of medication, behavioral, and
combined treatments on parents’ and children’s attributions for the behavior of children
with attention-deficit hyperactivity disorder. Journal of Consulting and Clinical
Psychology, 69, 67-76.

Johnston, C., & Mash, E]. (2001). Families of children with Attention-
Deficit/Hyperactivity Disorder: Review and recommendations for further research.
Clinical Child and Family Psychology Review, 4(3 ), 183-207.

Johnston, C. & Patenaude, R (1994). Parent attributions for inattentive-

overactive and oppositional-defiant child behaviors. Cognitive Therspy and Research,
l_8, 261-275.

Johnston, C., Pelham, W.E., & Murphy, HA. (1985). Peer relationships in
ADDH and normal children: A developmental analysis of peer and teacher ratings.

Journal of Abnormal Cm 1d Psychology, 13, 89-100.

Johnston, C., Reynolds, 8., Freeman, W.S., Geller, J. (1998). Assessing parent
attributions for child behavior using open-ended questions. Journal of Clinical Cm 1d

Psychology, 27, 87-97.

Kendall, P.C. & Wilcox, LE. (1980). Cogrritive-behavioral treatment for
irnpulsivity: Concrete versus conceptual training in non-self controlled problem children.

Journal of Consulting and Clinical Psychology, 48, 80-91.

Lahey, B.B. & Loeber, R. (1998). Attention-Deficit/Hyperactivity Disorder,
Oppositional Defiant Disorder, and Adult Antisocial Behavior: A life span perspective.
In D.M. Stoff, J. Breiling, & J .D. Maser (Eds), Handbook of Antisocial Behavior (pp.
51-74). New York: John Wiley & Sons.

Lazarus, R.S. & Folkman, S. (1984). Stress, sppraisal, and coping. New York:
Springer Publishing Co.

199

Lee, J. & Eccles, J .S. (1992). Teacher expectations: II. Construction and

reflection of student achievement. Journal of Personalig and Social Psflhology, 63(6):
947-961.

Lepper, MR. (1983). Extrinsic reward and intrinsic motivation: Implications for
the classroom. In J .M. Levine & M.C. Wang (Eds), Teacher and student percsptions:
Implications for learning (pp. 73-105). Hillsdale, NJ: Erlbaum.

Levy, S.R., & Dweck, CS. (1998). Trait- versus process-focused social
judgment. Social Coggp'tiop, 16(1), 151-172.

Licht, B.G., Kistrrer, J .A., Ozkaragoz, T., Shapiro, S., & Clausen, L. (1985).
Causal attributions of learning disabled children: Individual differences and their
implications for persistence. Journal of Educational Psychology, 77(2), 208-216.

Lin, E.H. & Peterson, C. (1990). Pessimistic explanatory style and response to

illness. Behavioral Research and Therspy, 28(3), 243-248.

Loeber, R., & Stouthamer-Loeber, M. (1986). Family factors as correlates and
predictors of juvenile conduct problems and delinquency. In M. Tonry & N. Morris
(Eds), Crime and Justice (V 01. 7, pp. 29-149). Chicago: University of Chicago Press.

 

Loney, J ., Whaley-Klahn, M.A., Kosier, T., & Conboy, J. (1983). Hyperactive
boys and their brothers at 21: Predictors of aggressive and antisocial outcomes. In K.T.

van Dusen & S.A. Mednick (Eds), Prospective studies of crime and delinquency
(pp.181-206). Boston: Kluwer-Niijhoff.

Mash, E.J., & Johnston, C. (1983). Parental perceptions of child behavior
problems, parenting self-esteem, and mothers’ reported stress in younger and older

hyperactive and normal children. Journal of Consulting and Clinical Psychology, 51, 86-
99.

McCubbin, H.I., & Patterson, J .M. (1983). The family stress process: The
double ABCX model of adjustment and adaptation. Social stress and the family. (eds).
Haworth Press Inc., pp. 7-37.

McGillicuddy-DeLisa, A.V. (1992). Parents’ beliefs and children’s personal-
social development. In I.E. Sigel, A.V. McGillicuddy-DeLisi, & J .J . Goodnow (Eds).

Parental Belief Systems: the psychological consgguences for children (2“d Edition), (pp.
115-142). Hillsdale, New Jersey: Lawrence Erlbaum Associates.

McGillicuddy-DeLisi, A.V. (1982). The relationship between parents’ beliefs
about development and family constellation, socioeconomic status, and parents’ teaching

strategies. In L.M. Laosa & I.E., Sigel (Eds), Families as learning environments for
children (pp.261-299). New York: Plenum.

200

Milich, R. (1994). The response of children with ADHD to failure: Ifat first you
don’t succeed, do you try, try, again? School Psghology Review, 23, 11-18.

Milich, R, Carlson, C.L., Pelham, W.E., & Licht, B.G. (1991). Effects of
methylphenidate on the persistence of ADHD boys following failure experiences.

Journal of Abnormal Child Psychology, 19, 519-536.

Milich, R., Licht, B.G., Murphy, D.A., & Pelham, W.E. (1989). Attention-deficit
hyperactivity disordered boys’ evaluations of and attributions for task performance on
medication versus placebo. Journal of Abnormal Psychology, 98, 280-284.

Milich, R., & Okazaki, M. (1991). An examination of learned helplessness
among attention-deficit hyperactivity disordered boys. Journal of Abnormal Child

Psychology, 19, 607-623.

Miller, SA. (1995). Parents’ attributions for their children’s behavior. Child
Developmeng, 66, 1557-1584.

Mischel, W., Zeiss, R., & Zeiss, A. (1974). Intemal-extemal control and
persistence: Validation and implications of the Stanford Preschool Intemal-Extemal

Scale. Joumpl of Personaligy & Social Psychology, 29(2), 265-278.

Morton, T.L. & Mann, B]. (1998). The relationship between parental
controlling behavior and perceptions of control of preadolescent children and

adolescents. Journal of Genetic Psychology, 159(4), 477-491.

The MTA Cooperative Group (1999). A 14-month randomized clinical trial of
treatment strategies for attention-deficit/hyperactivity disorder. Archives of Gener_a_l

Psychiafl, 56, 1073-1086.

Nigg, J .T. & Goldsmith, RH. (1998). Developmental psychopathology,
personality, and temperament: Reflections on recent behavioral genetics research.
Human Biology, 70 387-412.

 

Okagaki, L. (2001). Parental beliefs, parenting style, and children’s intellectual
development. In E.L. Grigorenko, & R.J. Stemberg (Eds). Family environment and

intellectual functioning: A life-span per_spective. (pp. 141-172). Mahwah, NJ:
Lawrence Erlbaum Associates, Inc.

O’Leary, K.D. (1980). Pills or skills for hyperactive children. Journal of
Applied Behavior Analfiis, 13, 191-204.

Overmeyer, S., & Taylor, E. (1999). Annotation: Principles of treatment for
hyperkinetic disorder: Practice approaches for the UK. J ourn_al of Child Psychology and

Psyhiafl, 40(8), 1147-1157.

201

Patterson, G. R (1996). Some characteristics of a developmental theory for early-
onset delinquency. In M.F. Lenzenweger & J.J. Haugaard (Eds), Frontiers of
Qevelppmental Psychopathology (pp. 81-124). New York: Oxford Press.

Pelharrr, W.E., Gnagy, E.M., Greiner, A.R., Hoza, B., Hinshaw, S.P., Swanson,
J .M., Simpson, S., Shapiro, C., Bukstein, O., Baron-Myak, C., & McBurnett, K. (2000).
Behavioral versus behavioral and pharmacological treatment in ADHD children attending
a summer treatment program. Journal of Abnormal Child Psyphology, 28, 507-525.

Pelham, W.E., Murphy, D.A., Vannatta, K., Milich, R., Licht, B.G., Gnagy, E.M.,
Greenslade, K.E., Greiner, A.R., & Vodde-Hamilton, M. (1992). Methylphenidate and
attributions in boys with attention deficit hyperactivity disorder. Journal of Consulting
and Clinical Psychology, 60, 282-292.

Peterson, C. (1988). Explanatory style as a risk factor for illness. Cognitive
Thsgpy & Research, 12(2), 119-132.

Peterson, C., & Barrett, LC. (1987). Explanatory style and academic

performance among university freshmen. Journal of Personalig and Social Psychology,
_5_3_, 603-607.

Peterson, C. & Bossio, L.M. (1991). Health and Cptimism. New York, NY:
Free Press.

Peterson, C., & Bunce, SC. (1997). Gender differences in personality correlates
of explanatory style. Personaligy and Individual Differences, 23, 63 9-646.

Peterson, C., Maier, S.F., & Seligman, M.E.P. (1993). Learned Helplessness: A
theory for the age of personal control. Oxford, New York: Oxford University Press.

Peterson, C. & Park, C. (1998). Learned helplessness and explanatory style. In
D.F. Baraone & M. Hersen (Eds). Advsnced Personalig. New York, NY: Plenun Press,
pp.287-3 10.

Peterson, C., Seligman, M.E.P., Yurko, KH., Martin, L.R., & Friedman, HS.
(1998). Catastrophizing and untimely death. Psychological Science, 9, 127-130.

Peterson, C., Semmel, A., Baeyer, C. Abramson, L.Y, Metalsky, G.I., &
Seligman, M.E.P. (1982). The Attributional Style Questionnaire. Cog'tive Th§_rr_tpy
and Research 6 287-300.

 

Peterson, C., & Vaidya, RS. (2001). Explanatory style, expectations, and
depressive symptoms. Personality & Individp_al Diffgrences, 31(7), 1217-1223.

Pearl, R.A. (1982). LD children’s attributions for success and failure: A
replication with a labeled LD sample. Learning Disability Ogapterly. 5. 173-176.

202

Pistermarr, S.J., McGrath, P., Firestone, P., & Goodman, J .T. (1989). Outcome
of parent-mediated treatment of preschoolers with Attention Deficit Disorder. Journal of
Consulting and Clinical Psychology, 57, 628-635.

Plomin, R., Nitz, K., & Rowe, DC. (1990). Behavioral genetics and aggressive
behavior in childhood. In M. Lewis & S.M. Miller (Eds), Handbook of developmental
psychopathology (pp. 119-133). New York: Plenum.

Podolski, C. & Nigg, J .T. (2001). Parental stress and coping in relation to child
ADHD severity and associated child disruptive behavior problems. Journal of Clinical

led Psychology, 30, 503-513.

Pollard, 8., Ward, E.M., & Barkley, R.A. (1983). The effects of parent training
and Ritalin on the parent-child interactions of hyperactive boys. Child and Family
Behavior Ther__apy, 54(4), 51-69.

Rappley, M.D., Gardiner, J .C., Jetton, J .R., & Houang, RT. (1995). The use of
methylphenidate in Michigan. Archives of pediatrics and adolescent medicine 149 675-
679.

 

Reid, M.K., & Borkowski, J .G. (1987). Causal attributions of hyperactive
children: Implications for teacher strategies and self-control. Journal of Educational
Psychology, 79, 296-307.

Reirners, T.M., Wacker, D.P., Derby, K.M., & Cooper, L.J. (1995). Relation
between parent attributions and the acceptability of behavioral treatments for their
children’s behavior problems. Behavioral Disorders. M3), 171-178.

Reynolds, C.R. & Kamphaus, RW. (1992). Behavioral assessment sgtem for
children manual. Circle Pines, MN: American Guidance Service.

Rhee, S.H., Waldman, I.D., Hay, D.A., & Levy, F. (1999). Sex differences in
genetic and environmental influences on DSM-HI-R attention-deficit/hyperactivity
disorder. Journal of Abnormal Psychology, 108(1), 24-41.

Rice, M.B., & Harris, GT. (1997). The treatment of adult offenders. In D.M.
Stoff, J. Breiling, & J .D. Maser (Eds). _I;I_apdbool_( of Antisocial Behavior pp. 425-435,
New York: John Wiley & Sons, Inc.

 

Richters, J ., Arnold, L., Abikoff, H., Conners, C., Greenhill, L., Hechtrnan, L.,
Hinshaw, S., Pelham, W., & Swanson, J. (1995). The National Institute of Mental
Health Collaborative Multisite Multimodal Treatment Study of Children with Attention-
Deficit Hyperactivity Disorder (MTA): I. Background and rationale. Journal of the
American Acsdemy of Child apd Adolescent Psychistry. 34. 987-1000.

203

Robin, A.R., & Foster, S. (1989). Negotiating parent-gqlgent conflict. New
York: Guildford Press.

Rosen, L.A., O’Leary, & Conway, G. (1995). The withdrawal of stimulant
medication for hyperactivity: Overcoming detrimental attributions. Behavior The_r_apy,
&, 538-544.

Rubin, RH. & Mills, R.S.L. (1992). Parents’ thoughts about children’s socially
adaptive and maladaptive behaviors: Stability, change, and individual differences. In
I.E. Siegel, A.V. McGillicuddy-DeLisi, & J .J . Goodnow (Eds), (pp. 41-69). Parental
Belief Swems: the psychological consguences for children. (2“d Edition). Hillsdale,

NJ: Lawrence Erlbaum.

Rubin, RH., Mills, R.S.L., & Rose-Krasnor, L. (1989). Maternal beliefs and
children’s social competence. In B.H. Schneider, G. Attili, J. Nadel, & R. P. Weissberg

(Eds), Social competence in developmental pegpective (pp. 313-331). Dordrecht, The
Netherlands: Kluwer Academic.

Rutter, M., & Giller, M. (1984). Juvenile delinquency: Trends and pmfives.
New York: Guildford Press.

Rydvalova-Mrug, S., Hoza, B., Pelham, W.E., & Gnagy, EM. (2000). Parental
attributions as predictors of ADHD children’s improvement. Presented at annual meeting
of the American Psychological Association. Washington, DC.

Schachar, R., Sandberg, S., & Rutter, M. (1986). Agreement between teachers’
ratings and observations of hyperactivity, inattentiveness, and defiance. Journal of

Abnormal led Psychology, 14, 331-345.

Shaffer, D., Fisher, P., Lucas, C.P., Dulcan, M.K., & Schwab-Stone, ME.
(2002). NIMH Diagnostic Interview Schedule for Children Version 1V (NIMH DISC-
IV): Description, differences from previous versions, and reliability of some common

diagnoses. Journal of the Amerifln Academy of Child and Adolescent Psychigy, 39(1),
28-38.

 

Shaffer, D., Schwab-Stone, M., Fisher, P.W., Cohen, R, et al. (1993). The
Diagnostic Interview Schedule for Children- Revised version (DISC-R): 1. Preparation,
field testing, interrater reliability, and acceptability. Journal of the American Academy of
Child and Adolescent Psychiag, 32(3), 643-650.

Sigel, LE. (1992). The belief-behavior connection: A resolvable dilemma? In
I.E. Sigel, A.V. McGillicuddy-DeLisi, & J .J . Goodnow (Eds). Parental Belief Systems:
the psychological consguences for children (2"d Edition), (pp. 433-456). Hillsdale,
New Jersey: Lawrence Erlbaum Associates.

 

204

Sigel, LE. (1983). Child development in research in learning and cognition in the
1980s: Continuities and discontinuities from the 19703. Merrill-Palmer Merly, 27,
247-171.

Skinner, EA. (1985). Determinants of mother sensitive and contingent
responsive behavior: the role of childrearing beliefs and socioeconomic status. In I.E.

Sigel (Ed). Parental Belief Systems: the psychological consguences for children.
(pp.52-8l). Hillsdale, NJ: Lawrence Erlbaum Associates.

Slep, A.M. & O’Leary, S.G. (1998). The effects of maternal attributions on
parenting: An experimental analysis. Journal of Family Psychology, 12 (2), 234-243.

Slomkowski, G, Klein, R.G., & Mannuzza, S. (1995). Is self-esteem and

important outcome in hyperactive children. Journal of Abnormal Child Pachology, 23,
303 -3 15.

Smith, A.M. & O’Leary, S.G. (1998). The effects of maternal attributions on
parenting: An experimental analysis. Journal of Family Psychology, 12, 234-243.

Smith, A.M. & O’Leary, S.G. (1995). Attributions and arousal as predictors of
maternal discipline. Co 'tive Th and Researc 19 459-471.

Snyder, J .J . (1991). Discipline as a mediator of the impact of maternal stress and
mood on child conduct problems. Develqpment and Psychopathology, 3, 263-276.

Sobol, M., Ashboume, D., Earn, B., & Cunningham, C. (1989). Parents’
attributions for achieving compliance from attention-deficit-disorder children. Journal of
Abnormal Child Psychology, 17, 359-369.

Steiger, J .H. (1980). Tests for comparing elements of a correlation matrix.

Psycholog'cal Bulletr_n,' 87, 245-251.

Stewart, M.A., Cummings, C., Singer, 8., deBlois, CD. (1981). The overlap
between hyperactivity and unsocialized aggressive children. Journal of Child Psychology
and Psychiagg, 22, 35-45.

Strassberg, Z. (1995). Social information processing in compliance situations by
mothers of behavior-problem boys. Child Developmeng, 66, 376-389.

Swanson, J .M. (1992). School based assessments and interventions for ADD
students. Irvine, CA: K.C. Publications.

Swanson, J .M., Kraemer, H.C., Hinshaw, S.P., Arnold, L.E., Conners, C.K.,
Abikoff, H.B., et al. (2001). Clinical relevance of the primary findings of the MT A:
Success rates on severity of ADHD and ODD symptoms at the end of treatment. Journal
of the Ameficap Agdemy of Chich and Adolescent Psychiatry. 40. 168-179.

 

205

Swanson, J .M., Lerner, M., March, J ., &. Gresham, F .M. (1999). Assessment and
intervention for Attention-Deficit Hyperactivity Disorder in the schools. Pediatric
Clinics of North Americsp, 46, 993-1009.

Sweeney, P.D., Anderson, K., & Bailey, S. (1986). Attributional style in

depression: A meta-analytic review. Personalig processes and individual differences,
59, 974-991 .

Szatrnari, P., Offord, D.R., & Boyle, M.H. (1989a). Ontario Child Health Study:
Prevalence of attention deficit disorder with hyperactivity. Journal of Child Psychology

and Psychisg, 30, 219-230.

Szatrnari, P., Offord, D.R., & Boyle, M.H. (1989b). Correlates, associated
impairments, and patterns of service utilization of children with attention deficit
disorders: Findings fi'om the Ontario Child Health Study. Journal of Child Psychology

and Psychiapy, 30, 205-217.

Tannock, R. (1998). Attention Deficit Hyperactivity Disorder: Advances in
cognitive, neurobiological, and genetic research. Journal of Child Psycholqu and

Psychiagy, 39 65-99.

Taylor, E., Sandberg, S., Thorley, G., & Giles, S. (1991). The Epidemiology of
Child Hyperactivig. Maudsely Monographs No 33. Oxford: Oxford University Press.

Teeter, P.A., & Semrud-Clikeman, M. (1995). Integrating neurobiological,
psychosocial, and behavioural paradigms: A transactional model for the study of ADHD.

Archives of Clinical Neuropsychology, 10(5), 433-461.

Thompson, R.J., & Gustafson, K.E. (1996). Adaptation to Chronic Illness.
Washington, D.C.: American Psychological Association.

Thompson, R.J., Gustafson, ICE, Hamlett, K.W., & Spock, A. (1992). Stress,
coping, and family functioning in the psychological adjustment of mothers of children
with cystic fibrosis. Journal of Pediatric Psychology, 17, 573-585.

Vitanza, S.A., & Guamaccia, CA. (1999). A model of psychological distress for
mothers of children with attention-deficit hyperactivity disorder. J ournpl of Child and
Family Studies. 8. 27-45.

Vitiello, B., Severe, J .B., Greenhill, L.L., Arnold, L.E., Abikoff, H.B., Bukstein,
O.G., Elliott, G.R., Hechtrnan, L., Jensen, P.S., Hinshaw, S.P., March, J .S., Newcorn,
J .H., Swanson, J .M., & Cantwell, DP. (2001). Methylphenidate dosages for children
with ADHD over time under controlled conditions: Lessons for the MTA. Journal of

Americjap Academy of Child and Adolescent Psychiapy, 40, 188-196.

 

206

Waldman, I.D., Rhee, S.H., Levy, F ., & Hay, DA. (1999). Causes of the overlap
among symptoms of ADHD, oppositional defiant disorder, and conduct disorder. In F.

Levy & D.A. Hay (Eds). Attentiop, genes, and ADHD. New York: Brunner-Routledge,
p.1 15-138.

Wechsler, D. (1991). Manual for the Wechsler Intelligence Scale for Childre_r_r_—
Third Edition (WISC-III). San Antonio, TX: Psychological Corportation.

Wechsler, D. (1992). Wechsler Individual Achievement Test. San Antonio, TX:
Psychological Corporation.

Weiner, B. (1980). A cognitive (attribution) — emotional — action model of
motivated behavior: An analysis of judgments of help-giving. Journal of Personalig and
Social Psychology, 39, 186-200.

Weiner, B., Graham, 8., & Chandler, C. (1982). Pity, anger, and guilt: An
attributional analysis. Personality Ed Social Psychology Bulletin, 8, 226-232.

 

Weiner, B., Russell, D., & Lerman, D. (197 8). Affective consequences of causal
ascriptions. In J .H. Harvey, W. Ickes, & R.F., Kidd (Eds), New directions in attribution
research (Vol. 2). Hillsdale, NJ: Erlbaum.

Weiss, G. & Hechtrnan, LT. (1993). Hyperactive children gr_own up (2“I ctr):
ADHD in children adolescents and adults. New York: Guildford Press.

 

Weiss, G., Hechtrnan, L., Milroy, T., & Perhnan, T. (1985). Psychiatric status of
hyperactives as adults: A controlled prospective lS-year follow up of 63 hyperactive
children. Journal of the American Academy of Child Psychiat_ry, 23, 213-220.

Weiss, M., Hechtrnan, L., & Weiss, G. (2000). ADHD in parents. Journal of
Americ_a_n Aca_dimy of Child and Adolescent Psychiagg, 39, 1059-1061.

Whalen, C.K., & Henker, B. (1999). The child with attention-deficit/hyperactivity
disorder in family contexts. In H.C. Quay and A.E. Hogan (Eds), Handbook of
Disruptive Behavior Disorders (pp. 139-155). New York: Kluwer Publishers.

Whalen, C.K. & Henker, B. (1991). Social impact of stimulant treatment for
hyperactive children. Journal of Learning Disabilities, 24(4), 231-241.

Whalen, C.K. & Henker, B. (1976). Psychostimulants and children: A review
and analysis. Psychological Bulletin, 83(6), 1113-1130.

Whalen, C.K., Henker, B., & Granger, DA. (1990). Social judgment processes
in hyperactive boys: effects of methylphenidate and comparisons with normal peers.
Journal of Abnormpl Child Psychology, 18, 297-316.

 

207

Williams, E]. (1959). The comparison of regression variables. Journal of the
Rog Statistical Sociegg, Series B, 21, 396-399.

Wright, B., Partridge, I., & Williams, C. (2000). Evidence and attribution:
Reflections upon the management of attention deficit hyperactivity disorder (ADHD).
Clinical Child Psychology and Psychiatly, 4, 626-636.

208

   

ll111111111111111111111111111111
3 1293 02504 4532