AN EMPIRICAL APPLICATION OF TEMPERAMENT THEORY
OF CHILD DEVELOPMENT TO HYPERACTIVITY
by
RICHARD C. LANGSDORF, B.A., M.S.
A DISSERTATION
IN
PSYCHOLOGY
Submitted to the Graduate Faculty of Texas Tech University in
Partial Fulfillment of the Requirements for
the Degree of
DOCTOR OF PHILOSOPHY
Approved
Accepted
/)4^n of/ i jhe (^rfeduate School
Augus t , 198 0
•D-
PUl
ACKNOWLEDGMENTS
I am deeply indebted to Professor Robert P. Anderson
for his constant encouragement and invaluable insights
throughout this research investigation. I am grateful to
Dr. James Goggin for encouraging me to look in new direc
tions for answers to old problems. I am also indebted to
the other members of my committee. Dr. Richard Haase, Dr.
Michael Bieber, Dr. James Clopton, and Dr. Robert Bell.
I would like to extend special thanks to fellow grad
uate student, John Simoneaux, who came through in a pinch,
by gathering the final data for this study for the absent
investigator.
I would also like to express my appreciation to Mrs.
Betty Snyder, Director of All Saints School, and Dr. George
Mecham, Director of Mecham School, for their support of this
project, and of course, to the many children and parents who
agreed to serve as subjects.
Most of all, I am indebted to my wife, Alene, whose
patient understanding, personal sacrifice, and unfailing
emotional support, helped m.e beyond measure.
11
TABLE OF CONTENTS
ACKNOWLEDGMENTS ii
LIST OF TABLES v
I. INTRODUCTION 1
Hyperactivity 6
Review of Etiological Theories 9
Temperament Theory of Child Development 2 0
Temperament and Behavioral
Disorders in Children 23
II. METHOD 4 0
Design 40
Subjects 44
Apparatus 4 8
Instrumentation 51
Procedure 5 9
Statistical Hypotheses 60
III. RESULTS 62
Hypothesis 1 62
Hypothesis 2 68
IV. DISCUSSION 78
Hypothesis 1 73
Hypothesis 2 80
Theoretical Implications 82
Clinical Implications 89
iii
Practical Application 91
Future Research Directions 94
V. suiyuyiARY AND CONCLUSIONS 9 7
REFERENCES 100
APPENDICES 107
A. LETTER TO PARENTS 107
B. PICTURE OF VIGILANCE TESTING APPARATUS 109
C. RAW SCORE DATA FOR TWO MEASURES OF ACTIVITY LEVEL FOR THREE COMPARISON GROUPS ill
D. BEHAVIORAL STYLE QUESTIONNAIRE 114
E. ZUKOW HYPERKINESIS RATING FORM 124
F. PARENT CONSENT FORM 127
IV
LIST OF TABLES
1. Comparison of Hyperactive, Nonhyperactive, and Control Groups on Subject Measures 4 5
2. Comparison of Three Groups on Parent Occupational Level 4 9
3. Pairwise Comparisons of Hyperactive (H), Nonhyperactive (NH), and Control (C) Groups on Three Zukow Rating Scale Factors 50
4. Means and Standard Deviations for Five Variables in Difficult Pattern for Clinical and Control Groups 64
5. Correlation Matrix for Temperament Variables 64
6. Summary of One-way ANOVA Results Between Clinical and Control Groups on Difficult Child Variables 65
7. Means and Standard Deviations for Three Extraneous Variables for the Clinical and Control Groups 66
8. Summary of One-way ANOVA Results Between Clinical and Control Groups on Three Extraneous Subject Variables 67
9. Means and Standard Deviations for Six Variables in Active Difficult Pattern for the Hyperactive and Nonhyperactive Groups 7 0
10. Summary of One-way ANOVA Results Between Hyperactive and Nonhyperactive Groups on Active Difficult Variables 70
11. Means and Standard Deviations for Three Extraneous Subject Variables for the Hyperactive and Nonhyperactive Groups 71
V
12. Summary of One-way ANOVA Results Between Hyperactive and Nonhyperactive Groups on Three Extraneous Subject Variables 71
13. Discriminant V^eights, Standard Deviations, and Products, for the Eight Variables Discriminating Hyperactive and Nonhyperactive Groups 73
14. Classification Matrix 75
15. Jackknifed Classification Matrix 7 6
VI
(>
CHAPTER I
INTRODUCTION
Hyperactivity is one of the most frequently diagnosed
childhood behavior disorders. It is commonly characterized
by motoric restlessness, poor attention span, distractibil-
ity, and a generally excessive activity level. Hyperactivity
is clearly a disorder of major significance. Kahn and
Gardner (1975), for example, documented the magnitude of the
problem, noting that approximately 40% of school-age children
referred to mental health clinics manifested the hyperkinetic
syndrome. The disorder may first be assessed during infancy
or early childhood and typically recedes as the adolescent
years approach (Werry, 1968). Despine the characteristic
decrease in hyperactive symptoms with increasing age, such
children often exhibit residual adjustment problems in later
years (Laufer, 1971).
A wide range of causal explanations for hyperactivity
v have been suggested. Organic etiologies include brain dys-
function (Luria, 1961) and genetic factors (Steward, 1970).
Psychogenetic explanations have included emotional disturb
ance (Friedland & Shilkret, 197 3) and disordered child-parent
relationships (Laybourne, 1976) . Other investigators have
assumed an interactionist viewpoint, emphasizing the inter
play between organismic and environmental factors (Williamson,
1
Anderson & Lundy, 198 0) . An interactional viewpoint was
investigated in the present study.
From the interactional perspective, both normal and
deviant child development results from an on-going dynamic
interaction between the child's individual attributes,
including abilities, motives and constitutional factors, and
the child's intra- and extra-familial environment. Rather
than positing a single causal factor to hyperactivity, the
interactive position emphasizes the importance of the dynamic
interplay of multiple factors which might contribute to and
maintain a child's hyperactive behavior (Williamson et al.,
1980) .
In their theoretical exposition of temperament and its.
relationship to child development, Thomas and Chess (1977)
and Thomas, Chess and Birch (1968,1970) offered a potentially
useful schema for identifying populations of children with a
high risk of developing behavioral disorders, including hyper
activity. According to Thomas and Chess (1977) each child
has an individual behavioral style of responding to, and
coping with, environmental expectations and demands. This
individual style defines the child's temperament and can be
observed from infancy onward. Temperament is not attributed
to genetic, physiological, psychological, or environmental
etiological factors, but rather is used as a phenomienal terrt
describing the behavioral style of the child. Nine
temperament qualities have been specified: Activity level,
rhythmicity of biological function*, initial approach-
withdrawal, adaptability, intensity, miood, persistence, dis-
tractibility, and sensory threshold. Both normal and patho
logical development is conceived as being determined by the
"dynamic interaction between a child's individual temperament
and the environmental experiences, expectations and demands
which the child encounters" (Thomas & Chess, 1977). Based on
data gathered from the New York Longitudinal Study, three
temperament constellations were delineated: the Difficult
child, Slow-To-Warm-Up child, and the Easy child. For
present purposes only the first cluster of characteristics
designated the Difficult child was considered. Of the nine
behavioral categories examined in this study, five were found
to be significantly related to this particular group of chil
dren. These include, irregularity in bodily functions,
intense reactivity, tendency to withdraw from new stimuli,
slow to adapt to changes in the environment, and a general
negativity of mood.
A most striking finding pertaining to this temperament
constellation was the fact that 70% of the Difficult children
developed behavioral problems in contrast to 18% of the Easy
children (Thomas, Chess & Birch, 1968). Thus, membership in
this former group appears to entail a high risk of behavioral
pathology. Thomas and Chess (1977) offered an explanation
4
for this finding based on their theory of temperament. A
child's temperament characteristics are never considered to
be the only cause of behavioral disturbances. Rather deviant
development is always the result of the interaction between
a child's individual temperament make-up and significant
environmental features. Certain situations and demands are
more likely to create excessive stress for a child with a
specific temperament. "For the Difficult child the stressful
demands were typically those of socialization, namely, the
demand for alteration of spontaneous responses and patterns
to conform to the rules of living of the family, the school,
the peer group, etc." (Thomas et al., 1977, p. 39). The
parents of a Difficult child must cope with the irregularity
and slowness to adapt which may strain the parent-child rela
tionship. Such children are a trial for most parents since
they require a high degree of consistency and tolerance in
their upbringing. The finding that 7 0% of this group
exhibited behavioral problems would suggest that these parents
are often unable to meet this child rearing challenge
successfully.
Several investigators have suggested a possible linkage
between the Difficult child temperament and hyperactivity
(Goggin, 1972; Henderson, Dahlin, Partridge & Engelsing, 1973,
1984; Ross & Ross, 1976; Steward & Olds, 1973). Henderson
et al. (1973) for example, hypothesized that hyperactivity
may make its first appearance in the preschool child but is
detectable from early infancy. "The tense, colicky, hyper
tonic infant seems to be the infantile precursor to the
hyperactive older child" (p. 625). They hypothesized that
hypersensitivity of the infant nervous system may predispose
it to easy arousability. Parents of such an infant may re
act with increasing tension and irritability that serves to
further increase the infant's arousal level. A negative
cycle is set in motion that results in a chronically stressed
parent-child relationship and sets the stage for hyperactiv
ity. Henderson et al. (1973) suggested the high activity
level, short attention and distractibility were secondary
symptoms and could be significantly modified by directly
addressing the stressed' parent-child relationship. The tense,
colicky, hypertonic infant is strikingly similar to Thomas
and Chess' Difficult child temperament pattern.
Thus, there is evidence that temperament theory m.ay be
able to make a significant contribution to the area of child
hood hyperactivity. The present study investigated the util
ity of the Thomas et al. (1968, 1970) theoretical formulations
in identifying and differentiating among populations of hyper
active and nonhyperactive children, and offered a conceptual
framev/ork for understanding the development and maintenance
of the hyperactive behavior.
Hyperactivity
prevalence
It has been estimated that 5,000,000 children in the
United States are currently labeled as hyperactive (Wender,
1973) . Estimates of the incidence of the disorder range from
three percent to ten percent of school children under 12
years of age. A report from the Department of Health, Educa
tion, and Welfare stated that three percent of all elementary
school children experience moderate to severe hyperkinetic
disorders. Huessy (1967), in a survey study of 300 second
grade children in rural Vermont, found a ten percent inci
dence rate for hyperactivity. Finally, approximately 4 0% of
all children referred to child guidance clinics are labeled
hyperactive (Kahn & Gardner, 1975). Hyperactivity is clearly
a problem of significant proportions.
The disorder may first be noticed during infancy or
early childhood (Werry, 1968). More typically the child
exhibiting hyperactive behavior comes to the attention of
mental health professionals after entering school since the
restless and distracted child has a disruptive effect on the
classroom.. Although the behavioral characteristics of hyper
activity diminish with increasing age such children frequently
exhibit residual adjustment problem.s in later years including
delinquency (Lewis, Sachs, Bella, Lewis & Heald, 1973), aca
demic failure (Rubin & Balow, 1971) and emotional
7
maladjustment (Minde, Weiss & Mendelson, 1972). Laufer
(1971) found that approximately one-third of a previously
diagnosed group of hyperactive children needed psychiatric
treatment during their adolescent years. Thus, these find
ings suggested hyperactivity may be associated with long-
term adjustment difficulties.
The Problem of Definition
Precisely what defines "hyperactivity" continues to be
an unsettled issue. Anderson (1977) distinguished between
monosymptomatic and multisymptomatic definitions.
Monosymptomatic definitions emphasize specific behav
ioral symptoms displayed by the child, such as level of
activity or attentional deficit. Chess (1960), for example,
offered the following monosymptomatic definition of the hyper
active child as one who "carries out activities at a higher
rate of speed than the average child, or who is constantly
in motion, or both." Likewise, Werry (1968) described "devel
opmental hyperactivity" as "a level of daily motor activity
which is clearly greater . . . than that occurring in chil
dren of simiilar sex, mental age, socioeconomic and cultural
background and which is not accompanied by clear evidence of
major central nervous system disorder or childhood psychosis
and which has been present consistently since the earliest
years of life." Both of these definitions focus on nhe
8
quantitative level of the behavior. In fact, the situ
ational appropriateness of the activity may be equally impor
tant. Stressing the value of the social context, Schmitt
et al. (1973) defined the disorder as an activity level
judged excessive when teachers and parents complain about it.
In contrast to monosymptomatic descriptions, multisymp-
tomatic definitions of hyperactivity stress clusters of
behaviors that form a clinical syndrome. The Diagnostic and
Statistical Manual of Mental Disorders (1968) described the
hyperkinetic reaction of children as a disorder characterized
by overactivity, restlessness, distractibility, and short
attention span. Cantwell (1975) offered a second multisymp-
tomatic definition that included the characteristics of
hyperactivity, impulsivity, distractibility and excitability.
To this list of symptoms other investigators have added poor
coordination and learning difficulties (Clements, 1966;
Laufer & Denoff, 1957) .
Senf (1976) underscored this lack of definitional uni
formity by observing that over 9 9 different symptoms have
been attributed to minimal brain dysfunction, a syndrome
often used interchangeably with hyperactivity. This defini
tional quagmire was resolved for the purposes of the present
study by selecting subjects for membership in the hyperactive
group based upon a set of multiple criteria including (1)
referral for hyperactivity evaluation, (2) parent ratings of
hyperactive behaviors and (3) behavioral measurement of the
child's gross motor activity level.
Review of Etiological Theories
A wide range of causal explanations for hyperactivity
have been suggested. The following discussion surveys repre
sentative models from the following three categories:
organic, which includes brain dysfunction, biochemical altera
tions and genetic factors; psychogenic, including emotional
disturbance, disordered child-parent relationships, and
pathonomic family situations; and multigenic, v/hich empha
sizes organism-environment interaction in the production of
hyperactivity.
Organic Models
From a historical perspective the hyper.kinetic syndrome
was first associated with an organic etiology. The 1918
epidemic of encepnalitis in the United States left children
who were stricken with the disease exhibiting the hyperactive
symptoms. Around the same time it was observed that early
head injury or insufficient oxygen during or shortly after
delivery often resulted in hyperkinetic behavior in the child.
Thus, the disorder came to be associated with the "brain
damage syndrome" in children (Steward, 1970). Since this
early conceptualization of the disorder, however, it has beer.
observed that most children diagnosed as hyperactive do not
10
have a history of brain trauma nor do the majority exhibit
neurological signs reflecting central nervous system
deviation (Douglas, 1972).
Luria (1961) has advanced a model of cerebral dysfunc-
tioning to explain hyperactivity. According to this theory,
in the normally developing child the mediating effect of
language serves to inhibit the motoric response system.
When there is early brain damage prior to the development of
speech, the verbal intellective functioning may subsequently
develop normally, while the volitional motor system remains
disrupted. One possible result of this disruption would be
a pathological deficit in the inhibitory processes, that is,
speech is not regulating motor responsiveness and the child •
displays an excessive level of activity.
Other explanations associated with the brain dysfunction
model are offered by Anderson (1963), who posited a lack of
adequate integration of various perceptual modalities result
ing in minimal brain damage, and by Laufer, Denhoff and
Solomons (1957) who observed that developmental histories of
hyperactive children often reveal evidence of brain trauma
within the first five years of life, leading these authors
to suspect diencephalon dysfunction.
Genetic factors have been cited by Stev/ard (1970) as a
possible causative factor of hyperactivity. Morrison and
Steward (1971) demonstrated an association between hyperactive
11
children and the frequency of hysteria, sociopathy, and
alcoholism found in the biological first and second degree
relatives of these children. In contrast, the adoptive
parents and relatives of the hyperactive children studied
did not reveal the same high frequency of alcoholism, socio
pathy and hysteria. The important assumption here was that
these three personality disorders may have a genetic basis
and the fact that hyperactivity v/as shovm to be associated
with them in these studies supported the genetic mode of
transmission for hyperactivity (Ross & Ross, 1976). The
polygenetic hypothesis postulated by Morrison and Steward
(1973) stated that the greater the number of individuals in
the family who are affected, the higher the risk component
in that family. Thus, if hyperactivity is linked with alco
holism for example, families where there were two or more
cases of hyperactivity would have a higher prevalence of
alcoholism than families that have only one hyperactive child.
Morrison and Steward found evidence to support this conten
tion. These investigators concluded, however, that while
these findings appear to support the polygenetic transmission
hypothesis, they do not rule out the equally plausible social-
environmental mode of transmission.
A biochemical theory advanced by V7ender (1975) proposed
that deficiencies of neurotransmitters, especially norepine
phrine in the ascending reticular activating systei?. (RAS) ,
12
was a critical factor in causing hyperactivity. The RAS is
a key region concerning consciousness and arousal and there
fore dysfunction in this system would be expected to impair
attentional behavior. The low noradrenaline level that
Wender has postulated was present in the hyperactive child
would decrease activity in the inhibitory system, thus
resulting in a high level of motoric activity in the child.
One advantage of this theory is that it explains how methyl-
phenidate may relieve hyperactive symptoms by increasing the
amount of norepinephrine available in the brain which would
lead to a more focused and better organized response to the
environment.
Another biochemical explanation of hyperactivity was
associated with children with confirmed lead poisoning.
David, Clark and Voeller (197 2) found that children with body-
lead concentrations below the level needed to produce overt
symptoms of toxicity did, nevertheless, exhibit hyperactive
patterns of behavior. However, it is not known whether the
involvement of lead in the behavioral disorder is primary,
contributory, or incidental.
In a recent exhaustive review of the research on the
organic, biochemical, and genetic research on hyperactivity,
Dubey (197 6) concluded that any assumption regarding the
presence of this disorder based on these factors was un
warranted without clear, unequivocal evidence. Moreover,
13
he added that such evidence was lacking in all but a minor
ity of cases of hyperactivity.
Psychogenic Models
Psychogenic explanations of hyperactivity have focused
on the interpersonal nature of the disorder, particularly
the adaptive or coping aspects of hyperactivity. Using the
term "interpersonal hyperactivity," Friedland and Shilkret
(1973) conceptualized the excessive activity levels of these
children as signifying a defensive operation whereby the
child attempts to cope with anxiety stemming from interper
sonal contact, particularly with adults. They found this is
a legitimate explanatory hypothesis v/hen the child has a
history of many traumatic experiences with adult figures.
The child is fearful of forming relationships with adults
who have in the past led to suffering and disappointment.
Thus, these authors conceived of hyperactivity as a defen
sive strategy for keeping others at a distance.
Utilizing a systems model, Laybourne (1976) underlined
the critical role that may be played by the family of the
hyperactive child. A pathonomic home environment may very
well maintain the symptomatic behavior if not, in fact, be
its cause. To support this contention Laybourne cited a
single case study where a child exhibiting severe hyperactive
behaviors was successfully treated with drugs only after the
family pathology was modified.
14
Scott (1973), viewing the disorder from a psychodynamic
perspective noted the relationship in children between hyper
activity and depression and interpreted the "constant rest
less, purposeless activity, impulsivity, short attention
span, and distractibility" as a defense against chronic
depression. The adult analog is the hypomanic individual
who attempts to fight off depression by over-compensating
with activity and cheerfulness.
In a paper appearing in the Psychoanalytic Study of the
Child, Mittleman (1957) mentioned a "chronic" form of hyper-
kinesis which may be a "late reaction to early restraint of
motility instituted because of somatic or psychosomatic
illness." She offered a case history to support this conten
tion where hyperkinetic, aggressive behavior "represented in
part a reaction and rebellion against the motoric limitations
in the early years of life" (p. 304).
These several conceptual explanations of hyperactivity
from the psychogenic perspective have in common the fact that
they have derived from clinical observations and have as yet
to be subjected to systematic empirical investigation.
Multigenic Models
Early models and theories of hyperactivity have tended
towards single factor explanations of this behavioral dis
order. There is growing appreciation, however, for multi-
factor models of hyperactivity which are more flexible and
15
conceptually able to entertain more than one causal explana
tion at a time. Multifactor models are particularly useful
given the heterogeneous quality of hyperactivity. The
following discussion will first examine non-interactional
multifactor models and then models which stress the interplay
between organismic and environmental factors.
Non-interactive Models
Non-interactional multifactor models are typically a
composite of single factor models and describe different
kinds of hyperactivity which have different kinds of etiolog
ical substrates. They often take the form of a taxonomy with
little attempt to integrate the various factors into a uni
tary conceptual framework.
In one of the earliest attempts to systematically dif
ferentiate types of hyperactivity, Chess (1960) , drawing from
a sample of 82 children seen in private practice and diag
nosed hyperactive, specified five etiologically-based cate
gories: (a) physiological or temperamental hyperactivity,
which is present from birth or early infancy and is accom
panied by an absence of a history of brain damage; (b) organic
brain disorder, where the child has a history of hyperactivity
and whose onset is subsequent to a known brain trauma; (c)
mental retardation without evidence of brain damage; (d) re
active and neurotic behavioral disorder, where hyperactivity
is one of a constellation of symptoms which expresses a
16
defensive reaction to environmental stress and which is
accompanied by a lack of evidence of hyperkinesis in early
infancy and no evidence of brain damage; (e) childhood
schizophrenia, where the total behavior pattern warrants a
diagnosis of schizophrenia and where hyperactivity appears
as one of many symptoms.
In contrast to Chess' straight forward taxonomio
approach, Marwitt and Stenner (1972) attempted to elaborate
two patterns of hyperactivity which reflect an underlying
and predominant biogenic or psychogenic substrate. These
authors emphasized the importance of specifying the exact
behavioral correlates of the hyperkinetic syndrome, since it
is from these that etiology is to be inferred. The charac
teristics of the Pattern I child is that (a) he is constantly
stimulated regardless of time, place, context; (b) shows vir
tually no capacity for sustained effort; (c) exhibits unstable
peer relationships; (d) is characterized by a lack of common-
sense; (e) displays perceptual motor problems; (f) is av/kward
and clumsy and (g) exhibits specific learning disabilities.
Pattern I supposedly reflects a primary biological etiology.
The Pattern n child's behavior includes (a) selectively show
ing excessive levels of activity; (b) being able to control
himself when the child deems it beneficial; and (c) capable,
if motivated, of sustained effort comparable to his age group.
Pattern II behavior signals emotional disturbance and anxiety
17
and the key precipitating factors are environmental. In the
second category hyperactivity is seen as a learned response,
acquired by the child to cope with the environment. The ex
istence of these two hypothesized patterns has yet to receive
empirical verification (Williamson, 1978).
Glennon and Nason (1974) proposed a model similar to the
dichotomy presented by Marwitt and Stenner (1972). It is
also a two-factor model consisting of primary and secondary
types of hyperkinesis. Primary hyperkinesis is attributed to
genetic factors, birth injuries, and infections. Secondary
hyperkinesis, on the other hand, is seen as resulting from
psychological or emotional factors, environmental or situ
ational factors, and suboptimal parenting.
Interaction Models
Multifactor interactional models are characterized by
their emphasis on a conceptual integration of the various
relevant organismic and environmental factors contributing
to the hyperactive pattern of behavior. Anderson (1970)
suggested an interactional model in which he stressed the
importance of both neurological and psychological factors
in the development of hyperactivity. While acknowledging
the possibility of different etiologies, he pointed out the
adaptive aspects as well as the secondary gains derived by
the child from his or her hyperactive behavior:
18
The hyperkinetic child may also learn to utilize his restlessness for achieving his own goals. He learns that being restless is a great source of attention and that it provides him with an excellent technique of disengagement from his responsibilities, (p. 145)
Recently Williamson, Anderson, and Lundy (198 0) proposed
an ecological model of hyperkinesis. According to this model,
behaviors characterizing hyperkinesis occur under certain
. antecedent conditions and result in situation-specific conse
quences. This model stressed the interplay between child and
environment. Various potential antecedents were categorized
into three primary domains, including the family, the academic,
and the physiological (i.e., internal) environments. Cogni
tive processes were also viewed as important antededent fac
tors referring to the hyperactive child's perception and
evaluation of the input from these three environments. How
these environments respond to the hyperactive child may range
from acceptance to intolerance. In the Williamson et al.
(1980) ecological model such responding refers to the
situation-specific consequences impacting on the child. These
consequences in turn, are cognitively processed by the child.
Depending upon the environmental input, the child's sense of
worth and self-esteem may be enhanced or deflated. The com
plexity of the ecological model lies in the emphasis upon a
continuous feedback loop between child and environment. For
this reason the model is likely to provide a more accurate
19
accounting of the development and maintenance of the hyper
activity than the single factor linear models discussed
earlier.
Another interactional theory was proposed by Henderson,
Dahlin, Partridge, and Engelsing (1973, 1974). These inves
tigators regarded the etiology of hyperactivity as a "chain
of events" with both primary and secondary sources of the
symptom behavior. Primary factors include characteristics
of normal individual genetic differences in reactivity and
arousability, plus intra-uterine, birth and postnatal traumas.
Secondary factors involve the responses the child elicits in
significant others in his environment. Thus, the tense,
colicky, hypertonic infant may be both tiresome and irritating
to the parent resulting in a mother-infant interaction which
spirals into states of increasing tension and difficulties
for both. Henderson et al. (1974) also introduced the possi
bility that factors which are usually secondary to the
infant's characteristics may in some cases initiate this nega
tive chain of events. In this case poor parental handling
may cause the originally calm child to become increasingly
tense and anxious which eventuates in behavior characteristic
of hyperactivity. Although not acknowledged by this group of
investigators, this model is highly similar to the tempera
ment theory of behavioral disorders in children developed by
Thomas, Chess and Birch in their series of publications
1 20
(Chess, 1968; Chess, Thomas, Rutter, & Birch, 1963; Thom.as &
Chess, 1972, 1977; Thomas, Chess & Birch, 1968, 1970). We
turn now to a detailed and extensive discussion of the tem
perament theory, particularly as it relates to the develop
ment of behavioral disorders in children.
Temperament Theory of Child Development
In their longitudinal study of temperament, Thomas and
Chess (1977) suggested that observable behavioral character
istics appear as early as infancy that persist over time and
are useful in defining personality types and understanding
subsequent behavioral disorders. In 1956 these investigators
launched their first longitudinal study which involved 13 6
children and their parents (Thomas, Chess, Birch, Hertzig &
Korn, 1963) . Later, other groups of children were studied
from different social class and cultural backgrounds (Hertzig,
Birch, Thomas & Mendez, 1968; Marcus, Thomas & Chess, 1969)
and with specific physical or mental disabilities (Chess &
Hassibi, 1970; Chess, Korn & Fernandez, 1971).
Temperament is defined by Thomas et al. (19 68) as
follows:
Temperament may best be viewed as a general term referring to the how of behavior. It differs from ability, which is concerned with the what and how well of behaving, and from motivation, which seeks to~account for why a person does what he is doing. When we refer to temperament, we are concerned with the way in which an individual behaves. Two children may each eat skillfully or throw a ball with
21
accuracy and have the same motives in so doing. Yet, they may differ with respect to the intensity with which they act, the rate at which they move, the mood which they express, the readiness with which they shift to a new activity, and the ease with which they will approach a new toy, situation, or playmate. Thus, temperament is the behavioral style of the individual child—the how rather than the what (abilities and content) or why (motivations) of behavior. Temperament is a phenomenologic term used to describe the characteristic tempo, rhythmicity, adaptability, energy expenditure, mood and focus of attention of a child, independently of the content of any specific behavior, (p. 4)
The identification of nine specific temperamental char
acteristics was achieved by an inductive analysis of the
behavioral protocols. These nine categories are:
1. Activity Level: The frequency and speed of movement
of the child; whether wiggling in the bath in early
infancy or walking or running in later developmen
tal stages.
2. Rhythmicity: Biological regularity or irregular
ity as seen in such functions as the sleep-wake
cycle and the timing of hunger and defecation.
3. Approach/Withdrawal: The immediate reaction of
the child to a new experience--such as a new person,
place, or object—in terms of his acceptance or re
jection.
4. Adaptability: Responses to new or altered situ
ations. Ease with which child adjusts to new
environments.
22
5. Threshold; The minimum strength of stimulus
required to engage the child's notice, without
regard to the positive or negative direction or
strength of reaction.
6. Intensity: The energy level of response, irrespec
tive of its quality or direction.
7. Quality of Mood: The amount of pleasant, joyful
and friendly behavior, as contrasted with unpleasant,
crying and unfriendly behavior.
8. Distractibility: The effectiveness of extraneous
environmental stimuli in interfering with or in
altering the direction of the ongoing behavior.
9. Attention Span and Persistence: Attention span
concerns the length of time a particular activity
is pursued by the child. Persistence refers to the
continuation of an activity in the face of obstacles
to the maintenance of the activity direction.
Three temperamental constellations of clinical signif
icance have been defined by qualitative analysis of the data
and factor analysis. These are the Difficult Child, the Easy
Child, and the Slow-To-VJarm-Up Child. The Difficult Child is
characterized by biological irregularity, predominance of neg
ative mood, high intensity of expressiveness, withdrawing
reactions to nev/ stimuli, and slow adaptability. The Easy
Child, in contrast, is usually described by the parents as a
23
"good baby." Such a child tends to be biologically highly
regular and predictable, has a predominantly positive mood,
expresses moods with mild or moderate energy level, approaches
new situations, and adapts quickly. The Slow-To-Warm-Up
Child temperament is characterized by a combination of nega
tive responses of mild intensity to new stimuli with slow
adaptability after repeated contact. In contrast to the dif
ficult children, these youngsters show mild emotional reac
tions, whether positive or negative, and show regularity of
biological functions. These children are often seen as shy
since they are likely to withdraw from new situations by
hiding behind mother and quietly moving apart from the group.
In the New York Longitudinal Study (NYLS) sample the Easy
children comprised about 4 0% of the total sample, the Slow-
To-Warm-Up group about 15%, and the Difficult children approx
imately 10% of the total sample. The remaining children
(35%) did not fit any of these three patterns. The focus of
the present investigation was on those children characterized
as Difficult.
Temperament and Behavioral Disorders in Children
Anterospective Method
An anterospective method of data collection v;as employed
in the New York study in order to avoid the distortions in
retrospective parental reports on the early developmental
24
histories of the children. Data were systematically gathered
at sequential age levels from early infancy to late childhood
Information included the nature of the child's individual
characteristics of functioning at home, in school, and in
standard test situations; on parental attitudes and child
care practices; on special environmental events and the
child's reactions to such events, and on intellectual func
tioning. Thus, utilization of the anterospective method
resulted in temperamental data being collected before the
child was perceived as a problem. .Thus, the data are uncon-
taminated by the distortions which are likely to accompany
retrospective histories gathered after the onset of the
behavioral disorder.
In the NYLS sample, 31% of the children developed behav
ioral problems serious enough to warrant psychiatric inter
vention. It should be noted that this figure is cumulative
over a ten-year period and at any given time percentages
would be lower.
The Difficult Child and Behavioral Disorders
A most striking finding of the New York project was that
the group of Difficult children, comprising only 10% of the
total sample, accounted for one quarter of the group develop
ing behavior disorders by the age of five years. Moreover,
within the group of Difficult children, 70% exhibited
25
behavioral disturbances (Thomas & Chess, 1977). Thus, member
ship in this group appears to entail a high risk of develop
ing psychopathology. It is important to recall that temper
ament theory postulates an interactive relationship between
the child and the environment. Thus, the child's pattern of
temperament alone is never the cause of behavioral disorder.
Rather, deviant•development as well as healthy growth, is
always the result of the dynamic interplay between a child's
individual temperamental make-up and significant environmen
tal features. For the Difficult child the stressful environ
mental demands are likely to be those of socialization
(Thomas & Chess, 1977). Initially it is the parents of the
'difficult' infant who must learn to cope with the child's
irregularity and slowness to adapt. Later, when the 'dif
ficult' youngster enters school, he or she is confronted with
a whole realm of novel adaptive tasks, social as well as in
tellective, which greatly increases the stress on the child
and enhances the potential for developing behavioral path
ology. With this interactional framework in mind, we shall
review the several studies which have examined this relation
ship between the child with the 'difficult' temperamental
pattern and the development of behavior disorders.
Two separate longitudinal studies by Chess and her asso
ciates (Chess & Hassibi, 1970; Chess et al., 1971) involving
children with intellectual or physical defects revealed that
26
even mild manifestations of the Difficult child temperament
were found to be associated with the development of behavior
disorders. In the first study (Chess & Hassibi, 1970) a
group of 52 mildly retarded children ages 5 to 11 were eval
uated. Fifty-nine percent were judged as having behavior
problems based on psychiatric examination. Of these 31
cases, 61% had three or more signs of the Difficult child
compared to the 21 children without disorders. In the group
without problems, only 14% had three or more Difficult child
signs.
In a companion study. Chess et al. (1971) investigated
243 children with a confirmed diagnosis of congenital rubella.
The children were between two and one-half and four years of
age when the study was initiated and suffered a variety of
physical handicaps, including visual deficits, hearing loss,
cardiac abnormalities and such problems as spasticity and
seizures. Again, nearly half (48%) exhibited behavior dis
orders and 4 0% of this group had four or five signs of the
Difficult child compared to only 14% of the group without
manifest psychopathology.
It is important to note that no statistically signif
icant differences were found in the frequency of the five
signs of the Difficult child between the original NYLS group
of youngsters and the groups of children with rubella or
mental retardation. Thus, differences between these three
27
groups in frequency of observed behavior problems cannot be
explained in terms of increasing prevalence of the Difficult
temperament pattern. Thomas and Chess (197 7) interpret the
greater incidence of disturbed children in the rubella and
mentally retarded groups as evidence that the 'difficult'
child with an intellectual or physical handicap is especially
vulnerable to developing behavioral disorders. The combina
tion of certain temperament traits with specific physical or
intellectual disabilities made for severe problems of adapta
tion and stress. For example, the deaf child with the tem
peramental tendency of initial withdrawal from new, unfamiliar
situations, because of major obstacles to communication, made
adaptation particularly difficult. Consequently, great under
standing, patience and empathy was required on the part of
parent or teacher for the child to cope successfully. This
example illustrated the critical importance of the child-
environment interaction in determining the degree of stress
experienced by the child.
In a recent longitudinal study, Mclnery and Chamberline
(1978) followed 118 infants who were divided into three groups
Difficult infants. Easy infants, and a control group. They
found the Difficult infant group had a higher frequency of
symptoms at two years of age and were described as more
aggressive and resistent than either the Easy infant group or
the total sample. Mclnerny and Chamberline concluded that
28
the data supported the hypothesis that the Difficult six-
month-old infant was likely to become the Difficult two-
year-old child.
In another study of 13 babies with colic, Carey (1972)
reported a significantly higher incidence of colic in
infants with the 'difficult' temperament than in the Easy
child group (p<.05).
In a study aimed at exploring mother-infant interaction
patterns, Feiring (1976) utilized an inventory designed to
measure infant temperament. In a group of 3 0 mothers and
infants, Feiring found that adaptive maternal behaviors,
defined as behaviors that have positive consequences for the
development of a secure infant-mother attachment, are related
to infant temperament. Specifically, Easy infants tended to
have mothers who were rated as high in adaptive maternal
behavior, while Difficult infants tended to have mothers
rated as low in adaptive behavior. Although the strength of
this association was not reported, the author concluded that
the results lend support to the hypothesis that stress in
the mother-infant relationship miay be related to the infant's
temperamental characteristics.
Effective prevention of behavioral disorders in children
depends upon the accurate identification of those children
most "at risk." Toward this end, Graham, Rutter, and George
(1973) studied the temperamental characteristics in a
29
population where there was a high risk of behavior disorders.
Each of the 60 three- to seven-year-old children in the
sample had at least one "mentally ill" parent, which pre-
siomably increased the likelihood of the offspring developing
psychological disorders. Approximately 41% of the group were
found to develop problems based on scores obtained on behav
ior questionnaires completed by parents and teachers. The
temperament categories used in this study were somewhat dif
ferent from those developed by Thomas et al. (1968). Essen
tially, the results pinpointed three temperament traits as
the main predictive categories of behavioral pathology.
These included: (a) low regularity, (b) lov/ malleability
(similar to adaptability in the NYLS project), and (c) low
fastidiousness (a characteristic not included in the New York
study) . Fastidiousness was defined as degree of "tolerance of
mess and dirt." The authors termed this triad, "adverse tem
peramental characteristics" and noted its similarity to the
Difficult Child constellation.
Finally, Cameron (1978) performed a fine grained anal
ysis on the original NYLS data. He computed a yearly "tem
perament risk" score for each child in the study. This score
was comprised of four behavioral categories similar to the
Difficult child constellation of traits. He found first-year
temperament scores predictive of mild cases of behavioral
disturbance for both sexes, although the temperament data
30
alone could not predict moderate-to-severe cases. The major
ity of cases were mild disorders. Nevertheless, these find
ings indicate that the relationship between temperament
pattern and psychopathology in children is likely to be
complex.
To summarize this section, the several empirical studies
that investigated the relationship between temperament pat
tern and behavioral disturbance generally provided support
for the hypothesis that the child's temperament is related
to the likelihood of developing some form of behavior dis
order in a predictive manner. On the other hand, as under
scored by Cameron's (1978) reanalysis of the original NYLS
data, the relationship is unlikely to reflect a simple one-
to-one correspondence.
The Relationship Between the Difficult Child Pattern and Hyperactivity
Before beginning this section it was important to recall
that hyperactivity is a category without precise boundaries.
With this caveat in mind hyperactivity was linked with the
temperament constellation referred to as the Difficult child.
Chess (1960) was perhaps the first to make the connec
tion between temperament and hyperactivity. Drawing from a
sample of 82 children seen in private practice and diagnosed
hyperactive. Chess delineated five etiologically-based cate
gories. One of these, termed "physiologic hyperactivity"
31
accounted for 42% of the entire sample and was defined by
(a) the presence of hyperactivity from birth or early
infancy and (b) the absence of any significant evidence of
brain damage.
A highly similar diagnostic category, "developmental
hyperactivity," was defined by Werry (1968) as hyperactive
behavior present since the earliest years of life and occur-
ring in the absence of major brain dysfunction or childhood
psychosis. Developmental hyperactivity is distinguished
from reactive or situational forms of the disorder by its
chronicity and an absence of clearly defined onset. It is
often associated with defects of attention, impulsiveness
and intensity of emotional reaction.
Henderson et al. (1973, 1974) hypothesized that the
tense, colicky, hypertonic infant may be the "infantile
precursor" to the hyperactive older child. Although no
research evidence was offered, these authors conjectured
that these temperament characteristics in the infant may be
due to the hypersensitivity of the nervous system predispos
ing it to easy arousability. Parents of such an infant may
react with increasing tension and irritability which further
increases the infant's arousal level setting in motion an
escalating stress cycle. Henderson et al. (1973) suggested
that the hyperactive child's symptomatic behavior, including
excessive motoric activity, short attention span, and
32
distractibility are secondary to a hypersensitive nervous
system and the resulting infant characteristics. The tense,
colicky, hypertonic infant was strikingly similar to the
Difficult child temperament pattern.
Steward and Olds (1973) also reported that many hyper
active children display characteristic behavior at an early
age. "They often are colicky and experience difficulty set
tling down to a particular formula. They do not sleep well
or regularly, and they are generally cranky, fussy babies.
They also exhibit inordinate amounts of energy . . ." (p.
26). Thus, these authors described the infant likely to
become the hyperactive child as biologically irregular, exhib
iting a general negative mood and high intensity of expres
siveness, as well as a high level of activity. Again, this
description closely approximates that of the energetic Dif
ficult infant.
At this point it is important to emphasize that the
Difficult child may or may not exhibit a high activity level,
the common denominator of all the various symiptomatic pat
terns designated hyperactivity. In connection with this
issue, Goggin (1972) suggested that when the Difficult child
has in addition a high activity level the behavioral descrip
tion is highly similar to the hyperkinetic syndrome. Tenta
tive support for this hypothesis was offered by McDevitt and
Carey (1978) who conducted a longitudinal study of 187
33
children to determine the stability of temperam.ent patterns
from infancy to early childhood. Mothers rated their child
at four to eight months of age and later between three and
seven years of age. One finding of this study was that in
dividuals who were difficult or intermediate high (a pattern
very similar to difficult) as infants and remained so in the
three- to seven-year period tended to be characterized by
elevated activity scores (15 of 17 children in this group
fell above the mean). Furthermore, these authors reported
that in a sample of 350 children, ages three to seven, used
in the standardization of their temperament questionnaire,
very high activity (more than one standard deviation above
the mean) was present in 28 of the 63 children classified
as difficult or intermediate high (£<.01). McDevitt and
Carey suggested a possible linkage between the configuration
of the difficult child with high activity and hyperactivity
by noting that "the active, difficult children undoubtedly
annoy their teachers more than most, and this may lead to
referrals resulting in the diagnosis of hyperactivity of
MED" (p. 337).
Only two studies have explicitly investigated the issue
of linkage between the Difficult child and hyperactivity.
Citing the paucity of longitudinal data on hyperactive chil
dren prior to school entrance, Campbell, Schleifer and Weiss
(1978) studied the relationship between maternal report of
34
problems in infancy and parent ratings of hyperactivity at
four and six years of age in two groups of children, a hyper
active and a control group. Infancy data included informa
tion regarding sleep problems, feeding difficulties, and
irritability, behaviors frequently associated with the
colicky infant. Mothers also rated their children at four
years of age on the Werry-Weiss-Peters Activity Scale which
provided a hyperactivity score and at six years of age on
the Connors Parent Questionnaire which is a more general
index of behavior problems and includes a hyperactivity
factor.
Campbell et al. (1978) found that maternal report of
sleeping problems during infancy were significantly corre
lated with ratings of hyperactivity at six years of age for
both the hyperactive (p<.05) and control (p<.01) groups.
Infant sleeping problems were also significantly correlated
with anxiety in the hyperactive group (p<.05) and conduct
problem.s (p<.05) and anxiety (p<.01) in the control group.
On the other hand, infant feeding problems and irritability
failed to correlate with ratings of hyperactivity. These
results can be considered as offering only modest support
for the temperament hypothesis that difficult infants have
a greater likelihood of developing behavior problems as
children.
<"•<•>
35
Unfortunately, this study suffered from several major
methodological problems. Definition of hyperactivity was
based solely on referral from an unspecified number of pedia
tricians who identified these children as "overactive" and
"difficult to manage." Also, attrition rate over the three-
year period exceeded 50% for the hyperactive group (15 out
of 28) and was likewise high for the control group (16 out
of 26), introducing the possibility of selective subject
mortality which may have biased the results in an undeter
mined manner. Furthermore, the use of retrospective methods
of data gathering are subject to distortion of recall
(Hetherington, 1972) which would be particularly critical
in this study. While the cumulative effect of' these method
ological shortcomings cannot be determined, caution in the
interpretation of the outcome certainly is warranted.
In the second study investigating the relationship
between temperament and hyperactivity, Lambert and V7indmiller
(1977) obtained interview data from 327 parents in a longi
tudinal study of elementary school children. The sample was
divided into four groups: hyperactive (N=35), low achievers
(N=55), school adjustment problems (N=126), and a "normal"
control group (N=lll). The investigators wanted to know
whether the Difficult child corresponded to the hyperactive
child and whether these temperament traits were present from
early infancy and childhood. They employed a standardized
tmm
36
interview format which included 17 questions gauged to
assess the nine temperament traits of Thomas et al. (1970).
The scores from the nine dimensions were subjected to factor
analysis which yielded six factors. The factors v/ere
labeled attention span, threshold level, activity level, dis
tractibility, approach/withdrawal, and rhythmicity/
adaptability.
The results showed that the hyperactive group had the
most extreme scores on all six temperament dimensions. How
ever, this difference was significant only on two categories,
distractibility and rhythmicity/adaptibility. Thus, these
findings would tend to suggest the Difficult child pattern
is not an efficacious predictor of hyperactivity. On the
other hand, on the basis of serious methodological short
comings, there is reason to believe the hypothesized rela
tionship has not been adequately tested. For example,
Lambert and Windmiller (1977) choose to develop their own
set of interview questions rather than work with those of
the original temperament researchers. While there appears
to be satisfactory correspondence with the original nine
factor analytic trait dimensions, the 17 questions used do
not equitably represent the Thomas et al. (197 0) temperament
categories. Specifically, while certain categories are
assessed v/ith three questions, others are measured with a
single question, resulting in a grossly uneven distribution
37
of the 17 questions across the nine item categories. Fur
thermore, the very unequal numbers of children in the four
groups, especially the relatively small number in the hyper
activity sample, may be responsible for the weak effects
obtained. In summary, whether or not Lambert and Windmiller
adequately tested the hypothesized relationship between the
Difficult child temperament pattern and hyperactivity is
open to serious question.
To summarize this section, the possible association be
tween the child with the difficult temperament and hyper
activity has been suggested by several investigators
beginning with Chess (1960) under the category of "physio
logical hyperactivity." However, only two recent studies
(Campbell et al., 1978; Lambert & Windmiller, 1977) have
explicitly examined this issue of a possible linkage between
the Difficult child and the hyperactive child. The Campbell
et al. study produced only lukewarm support for the hypoth
esis that the difficult infant in the forerunner of the
hyperactive child, finding a significant correspondence
between material report of infant sleep difficulties and
ratings of hyperactivity during childhood. The findings of
Lambert and Windmiller (1977) suggested only a weak linkage
between the Difficult child pattern and hyperactivity. How
ever, it was noted that both studies contained serious
methodological problems and thus there is reason to believe
the hypothesized relationship was not adequately tested.
38
In summary, research has shown that certain early trait
patterns correspond to the development of behavioral dis
orders in children (e.g., Thomas et al., 1968). A possible
relationship between the Difficult child pattern and hyper
activity also has been suggested, although not adequately
tested. The intent of the present study was to empirically
investigate the relationship between specific temperament
dimensions and behavioral disorders in children, especially
hyperactivity.
The research hypotheses were as follows:
1. There is a specific temperament trait pattern
which can differentiate a population of children
with known behavioral disorders from a popuation
of children without behavioral disorders. Spe
cifically, disturbed children in both the hyperactive
and nonhyperactive groups are more likely to be char
acterized by the Difficult Child pattern (arhyth-
micity, withdrawal, slow adaptability, intense
reactivity, negativity) than children in the control
group.
2. There is a specific temperament trait configuration
which can differentiate a population of hyperactive
children from a population of children not exhibiting
hyperactivity. Specifically, hyperactive children
are more likely to be characterized by the
39
temperament pattern of the Active Difficult Child
(high activity, arhythmicity, withdrawal, slow
adaptability, intense reactivity, negativity) than
children in either the nonhyperactive clinical
group or the control group.
•i
CHAPTER II
METHOD
The purpose of the present chapter is to describe the
methodological procedures utilized in the investigation of
the research hypotheses. The following topics are discussed:
(a) design, (b) subject characteristics, (c) apparatus, (d)
instrumentation, and (e) procedures. The chapter concludes
with a statement of the statistical hypotheses under
investigation.
Design
The study employed an ex post facto research design.
Briefly, this involved two independent steps:. (a) the assign
ment of each subject to the appropriate clinical group, i.e.,
hyperactive or nonhyperactive, or the control group, and (b)
the assessment of the subject's temperament.
The major purpose of the study was to investigate cer
tain predicted relationships between temperament and emo
tional/behavioral disorders in children. Therefore, a priori
membership in one of the three groups constituted the cri-
terion variable. The predictor variables, i.e., the variables
which predict subject's group membership, are the temperament
dimensions defined by Thomas et al. (1968, 1970). Temperament
was measured by means of the Behavioral Style Questionnaire
(McDevitt & Carey, 1978) , an instrument designed to measure
40
41
the nine temperament traits in children three to seven years
of age (see Instrumentation section).
Determiination of Group Membership
The three groups were: (a) hyperactive clinical, (b)
nonhyperactive clinical, and (c) control. Determination of
hyperactive and nonhyperactive clinical status was based on
the following three criteria:
1. Referral problem(s).
2. Subject's gross motor activity level recorded dur
ing a continuous performance task.
3. Mother's perception of subject as being or not
being "unusually hyperactive."
Specifically, subjects were assigned to the hyperactive
clinical group if they (a) were referred for evaluation of
possible hyperactivity by school personnel, physician, psy
chologist, or other health professional; (b) obtained a
standardized Total Activity Score above the mean (X=0.0)
during the Vigilance Task (see Apparatus section), and (c)
received a maternal rating of "unusually hyperactive" on a
checklist for children's problem behaviors (see Instrumenta
tion section).
Assignment to the nonhyperactive clinical group was
predicated upon (a) referral for evaluation for nonhyper
active behavioral or emotional problems by school personnel,
physician, psychologist, or other health professional, (b)
42
achieving a standardized Total Activity Score below the mean
during the Vigilance Task, and (c) received a maternal rating
of not "unusually hyperactive" on the problem checklist. The
following is a partial listing of the problems presented in
the nonhyperactive clinical group: temper tantrums, aggres
siveness, academic underachievement, chronic lying, sleep
difficulty, excessive shyness, learning disability, compul
sive behaviors and inordinate rigidity.
A given subject had to fulfill all three criteria in
order to be assigned to the appropriate clinical group. This
stipulation increased the relative homogeneity of the sub
jects within each clinical group with regard to the presence
or absence of the categorical variable "hyperactivity."
The stipulation that all three criteria had to be met in
order for a given subject to be assigned to one of the two
clinical groups, resulted in many potential subjects being
discarded from the study. This was especially true for the
nonhyperactive clinical group subjects. Of the potential
subjects who were eventually discarded, nearly all had been
referred for nonhyperactive problems. The measured activity
level of these children place many of them in the "hyper
active" range and they had to be dropped from the study. On
the other hand, the majority of the hyperactive group sub
jects had been referred after extensive interdisciplinary
evaluation in the Hyperactivity Clinic at the Texas Tech
University School of Medicine. Thus, the referrals from this
Clinic had a high a priori probability of meeting the cri
teria for inclusion in the hyperactive group.
A non-clinical control group was gathered from the stu
dent bodies of two private elementary schools. A letter
(Appendix A) was sent to the parents of potential subjects in
each school asking if they would be willing to participate in
a research study by completing a 100 item questionnaire and a
28 item checklist on their child. They were also asked to
grant permission for their child to be given two tests as
part of the same research project. Control subjects were
administered the same tests as clinical group subjects, i.e.
an individual intelligence test and Vigilance Task. At one
school the director elected to select students whose parents
he felt would be likely to participate in the study. Of the
15 subjects solicited, 14 (93%) participated. At the second
school, the investigator randomly selected 25 potential sub
jects from a student directory. Eighteen of those parents
(72%) agreed to participate in the project, although one
child refused to complete the Vigilance Task and therefore
could not be included in the study. The final combined con
trol group consisted of 31 subjects with a 77% overall
response rate.
44
Subjects
Subjects were 71 children, ages 4 years, 7 months to 7
years, 11 months. All subjects were residing in the West
Texas region with the majority living in the city of Lubbock,
Texas. Specific criteria pertaining to selection and assign
ment to the groups is detailed in the Design section.
The three groups of subjects are compared on the charac
teristics of age, intelligence test scores. Total Activity
scores and Total Error scores (Table 1). Mean age in months
for the three groups was highly similar with the hyperactive
group showing somewhat greater age range.
All children were administered an individual intelli
gence test. When the groups are compared on this character
istic, considerable discrepancy was evident between the
groups, with the hyperactive group obtaining the lowest
scores (X=96) and the control group the highest (X=120).
The elevated control group mean was a reflection of this
atypical subject sample which was gathered from an academ
ically accelerated private school population. The relative
elevation of the nonhyperactive clinical group mean (X=110)
may likewise be due to this same selection bias since 40%
of the subjects in this group were also attending private
schools. In contrast, all school-age subjects in the hyper
active group attended public schools.
TABLE 1.—Comparison of Hyperactive, Nonhyperactive, and Control Groups on Subject Measures
45
Hyperactive group (N=20)
mean
standard dev.
range
Nonhyperactive group (N=20)
mean
standard dev.
range
Control group (N=31)
mean
standard dev.
range
Age in months
81.80
10.70
54.95
82.15
8.36
61-94
79.70
8.06
66-93
12 a scores
96.50
13.55
80-137
110.60
12.13
89-142
120.41
12.63
94-155
Total activity scores
1.810
(1.020)
+0.084 to +4.028
-1.062
(0.532)
-0.050 to -1.726
-0.524
(1.075)
-1.726 to +1.828
Total error scores
73.60
59.76
28-293
28.40
18.29
2-85
18.41
14.08
0-58
^The Slossen Intelligence Test was used for 62 subjects. The WISC-R was administered to 9 subjects (5 in the hyperactive group and 4 in the nonhyperactive clinical group).
^Expressed as standard scores with X=0.0, S.D.=i.O, for total
sample (N=71).
46
As noted in the Design section, subject assignment to
either hyperactive or nonhyperactive clinical groups was, in
part, based on earned Total Activity score, a standardized
composite score, with a mean equal to zero and a standard
deviation equal to one. The two clinical groups are clearly
differentiated on this characteristic (Table 1), with all of
the subjects in the hyperactive group falling above the mean
while all nonhyperactive clinical subjects obtained scores
below the mean. There was no overlap between the two groups
on this characteristic. This was a criteria for inclusion
in the study and therefore potential clinical group subjects
who did not fulfill this requirement were not eligible and
had to be discarded. On the other hand, the control subjects
obtained a range of scores overlapping with both clinical
groups.
The three groups were compared on Total Error scores
(Table 1). Briefly, this was a composite error score earned
on the continuous performance test (see Apparatus section).
The score provided a measure of the subject's ability to
sustain a task-centered response set. A high error score
was equated with a greater attentional deficit. The hyper
active group exhibited a significantly greater attentional
deficit when compared with the nonhyperactive clinical group
(t=3.23, df=40, p<.01) and the control group (t=4.96, df=51,
p<.001). The difference between the nonhyperactive clinical
and control groups was less striking (t=2.20, df=51, p<.05).
9
47
Fifty-nine percent of the total sample were boys and 41%
were girls. The unequal representation of boyS and girls
was apparent in the hyperactive group (boys, 7 0%; girls, 30%)
and the nonhyperactive clinical group (boys, 60%; girls, 40%).
In the control group, on the other hand, there was approxi
mately equal representation with 16 boys and 15 girls.
Ninety-six percent of the sample were classified as
Anglo. There were only three non-Anglo children—one Oriental
and two Mexican Americans. All three non-Anglo subjects
belonged to the nonhyperactive clinical group.
Family status was defined in terms of whether or not the
child was residing with both biological parents. Seventy-two
percent of the subjects in the total sample were living with
both natural parents. The percentage of non-intact biolog
ical families increased sharply in the two clinical groups
(40% and 45% for the hyperactive and nonhyperactive groups,
respectively). Only 9.6% of the control group children came
from non-intact families.
An approximation of the socioeconomic status of the
child's family was estimated using an occupational hierarchy
based on Hollingshead's Two Factor Index of Social Position
as modified by Myers & Bean (1968). The hierarchy was pre
mised on the assumption that different occupations are valued
differently by members of society. For present purposes,
however, this schema offered a convenient means of ranking
3
9
dTM
48
subject's socioeconomic level as estimated by parental occu
pation. The seven position hierarchy ranges from the least
(7) to the most (1) prestigious occupations. The hierarchy
includes: (1) executives of large concerns and major profes
sionals; (2) managers and proprietors of medium concerns and
minor professionals; (3) administrative personnel of large
concerns, owners of small independent businesses and semi- .*
professionals; (4) owners of little businesses, clerical and 'i c
sales workers, and technicians; (5) skilled workers; (6) -«i '
semiskilled workers; and (7) unskilled workers. » 9
The occupational levels of the parents of the children
in the two clinical groups tended to distribute normally
across the seven level hierarchy (Table 2). In contrast,
the occupational levels of parents of control group children
clustered at the upper end of the hierarchy suggesting that
as a group, the control subjects were from a higher socio
economic group than the clinical subjects.
Apparatus
The Vigilance Task was employed primarily to differen
tiate hyperactive from nonhyperactive disturbed children.
Subject's level of gross motor activity was measured during
the task and constituted one of the three operational defi
nitions of hyperactivity.
49
TABLE 2 . — C o m p a r i s o n of Three Groups on P a r e n t O c c u p a t i o n a l L e v e l
Hyperact ive group (N=20)
>
1
2 10.0
r-l 3 10.0
0 4 40.0 •H
H 5 25.0 o 6 15.0 o ^ 7
Nonhyperactive group (N=
15.0
25.0
0.0
30.0
15.0
10.0
5.00
20) Control group
(N=31)
9.7
35.5
45.1
9-7
SI a
9 100.0 100.0 100.0 tf
The vigilance Task has been described in detail by
Anderson, Halcomb, and Doyle (1973). Basically, the task
requires children to attend to a series of visual events
occurring over time. A random sequence of visual signals
is typically interspersed within the events which the sub
ject must identify. A picture of the apparatus is contained
in Appendix B.
While sitting at a console, approximately 1.2m x 1.2m,
the subject observes a pair of flashing lights. The lights
are presented at two second intervals in random combinations
of red-red, green-green, or red-green, each with .2 second
duration. When the red-green combination appears, the sub
ject is to press a button mounted on a bicycle handlebar
grip. If the subject responds to the red-green combination
••n
0*n
50
by pressing the button, the response is scored as correct.
Responses to other combinations (i.e., red-red, green-green)
are scored as errors.
A total of 60 red-green correct detections are pre
sented within a 3 0-minute period, resulting in an overall
sum of 900 combinations within the testing session. The •-*
entire sequence of flashing lights and data recording is *
controlled by a digital control system. :j
In the present study two scores were computed for each
child, the total number of correct detections and the total
number of errors. By subtracting the number of correct
responses from the total possible correct responses and sum
ming this value with the number of incorrect responses, a
value designated Total Errors, was computed. Mean Total
Errors computed for each group are presented in Table 1 .
During the Vigilance Task the child sat on a stabi-
limetric chair, wired with mercury switches, that register
small movements in backward-forward and side-to-side direc
tions. A cumulative movement count was obtained at the end
of the experimental run. Foshee (1958), employing a similar
measuring device, reported high test-retest reliability on
successive days (r=.95).
In the present study, the stabilimetric chair provided
an "activity count," indicating the child's level of motoric
activity during the Vigilance Task. Since a restless child
51
may leave the chair during the task it was necessary to
adjust the measure to acconmiodate out-of-chair behavior.
Out-of-chair time was recorded with a stopwatch by the
experimenter. Each child earned two activity scores, a
stabilimetric count and a time-out-of-chair count. The two
scores were converted to standard scores and summed, yield
ing a composite standardized Total Activity score. The Total
Activity score results are shown for the three groups in
Table 1. Raw score values for both stabilimetric chair
count and time-out-of-chair count for each subject are listed
in Appendix C. For the total sample these two measures of
activity are relatively independent of one another (r=.ll).
On the other hand, for the hyperactive group, a significant
inverse relationship was found between the two measures
(r=-.66). Restless children tended to have high scores on
one measure or the other, but not both.
Instrumentation
Mothers of subjects in all three groups were requested
to complete the Behavioral Style Questionnaire (Appendix D)
and the Zukow Hyperactivity Rating Scale (Appendix E). Fur
thermore, all children in the study were administered an
individual intelligence test and the Vigilance Task.
Temperament Questionnaire
The Behavioral Style Questionnaire (BSQ) was developed
by McDevitt and Carey (1978) for determining temperamental
o SI
9
52
characteristics in young children. The 100 item instrument
was based on the theoretical work of Thomas et al. (1968,
197 0) and generates a score for each of the nine temperament
categories. The number of items per category range from 9
(Rhythmicity) to 13 (Activity). A six-point rating scale
from "almost never" to "amost always" is employed and the
parent is instructed to (a) consider only the child's recent
and current behavior; (b) to judge on their own impression
of the child, and (c) to consider each description indepen
dently. Each item contains a description of a child's behav
ior within a specific context. For example, "How often does
the child fall asleep as soon as he/she is put to bed?"
represents the rhythmicity category with respect to pattern
of sleeping. Behaviors indicative of both high and low
standing within each temperament category are included.
Items are arranged randomly as to content area and category
in order to avoid a response set by the rater.
Validity: Items were based on published description of
behaviors associated with each of the nine dimensions of
Thomas et al. (1963, 1968, 1970) and other behaviors judged
logically consistent with the category definitions. Initial
item selection was determined by inter-judge agreement of
category designation. Eight judges familiar with the NYLS
research assigned each item in the pool to one of the nine
categories. Only items correctly placed by five of the
•I
9
9r
53
eight judges were included in the pretest phase of this
instrument. This preliminary version was pretested on 53
kindergarden and first grade children and items correlating
poorly (r<_.30) and items correlating with several different
categories were revised or eliminated.
Standardization: The resulting instrument, called the .
Behavioral Style Questionnaire, was sent to 369 parents of * •I
3-7-year-old children in a pediatric practice and 350 ques- .J e.
tionnaires were returned (95%). There were 175 males and ::
17 5 females. The children were predominantly white and
from middle class families.
Test-retest reliability: Fifty-five of the 350 respon
dents were selected randomly for retest one month later with
a return rate of 96%. Test-retest reliability (N=53) ranged
from .67 to .94 for the nine categories. Total score test-
retest reliability was .89. Reliability for the six cate
gories of particular interest to the hypotheses under
investigation are as follows: activity (.93), rhythmicity
(.80), approach (.94), adaptability (.85), intensity (.75),
and mood (.87).
Internal consistency reliability: Based on the total
sample (N=350) an index of the homogeneity of the items
within each category was determined yielding alpha coeffi
cients ranging from .47 to .80. Internal consistency for the
overall instrument was .84. Alpha coefficients for the six
54
categories of interest in the present study are: activity "
(.76), rhythmicity (.48), approach (.80), adaptability (.72),
intensity (.71) and mood (.66).
There were few age or sex by category correlations and
those that were found that are relevant to the hypotheses
under investigation accounted for 3% or less of the variance.
Specifically, older children were rated as more rhythmic <
(r=-.18) than younger ones and males were found to be more -j
rhythmic (r=-.13) and more active (r=.12) than females. Mi
5''
Thus, for the most part, the items comprising the BSQ are J I
independent of age and sex of the child.
Relationship to the NYLS findings: In order to deter
mine the prevalence of the NYLS diagnostic groups within the
BSQ standardization sample, a person cluster analysis was
performed. Of the children in this sample 67% were included
in one of the three diagnostic clusters, compared to 65% in
the NYLS. The percentage breakdown by pattern of Difficult,
Easy, and Slow-to-Warm-Up was 18%, 32%, and 16% respectively,
compared to 10%, 3 5%, and 15% in the NYLS sample. Thus, with
the exception of the increase in the prevalence of the Diffi
cult cluster in the BSQ sample, the percentages are highly
similar lending credence to the assumption that the BSQ is
measuring the same temperament dimensions described by Thomas
et al. (1963, 1968, 1970).
55
One notable exception to this similarity between the
NYLS and BSQ samples is that the Difficult child constella
tion tended to be associated with high activity (p<.05) in
the latter sample, whereas in the original NYLS group the
Difficult children showed no consistent pattern .for activity.
It was this unexpected finding that prompted Carey and
McDevitt (1978) to conjecture that this subgroup.of Difficult
children (i.e., difficult pattern plus high activity level) i i
may be identified as hyperactive. This, of course, was the
second hypothesis investigated in the present study. f
Rating Scale for Hyperkinesis
The Rating Scale for Hyperkinesis (Zukow et al., 1978)
consists of 28 behavioral items in a forced choice format,
yielding the following three factor clusters: Excitability,
Motor Coordination, and Directed Attention. The first factor.
Excitability, is concerned with emotional pitch and tempo.
As Zukow et al. (1978) note, the items defining the Excitabil
ity factor are most similar to those characteristics of hyper
activity as described by others. The second factor. Motor
Coordination, refers exclusively to motor coordination, and
the items tap poor coordination, delays in walking, speech,
etc. The third factor. Directed Attention, refers to sus
tained participation in goal-directed behaviors.
The rating scale was developed on a total sample of 160
children, ages 2-1/2 to 11, with approximately equal
:]»
9
56
distribution of subjects among hyperactive and control
groups. The children were predominantly from middle income
families. Inclusion in the hyperactive group was based upon
careful screening which involved the judgement of a Pedia
trician who gathered data on medical and family history plus
behavioral evidence reported by parents and teachers. Fur
thermore, the child was assigned to the hyperactive group
only if there was consistent evidence of behavioral disturb- :{
ance at home and at school along with a presistent learning
disorder associated with an inappropriately high activity J
level. Evidence of overt neurological involvement, sensory
disability, and mental retardation were grounds for exclusion.
The control sample was drawn from children attending public
and private schools, after a screening for hyperkinetic
behavior problems.
The three factors (excitability, motor coordination,
and directed attention) as well as a total score were found
to discriminate among hyperactive and control status at sta
tistically significant levels (p<.001). Internal consistency,
as measured by coefficient alpha, yielded reliabilities of
.84, .74, and .82 for the three factors, excitability, motor
coordination, and directed attention, respectively. Although
no test-retest reliability data is offered, the authors did
cross-validate their instrument by comparing the scores of a
new hyperactive sample (N=26) with the original control group
mm
1
57
and again obtained significant differences on all factors
(p<.001).
The rating scale served two purposes in the present
study. First, the response of the child's m.other to the
first item (Unusually hyperactive: Home School Both
No) was a criteria for determining the parent perception of
the clinical subject as hyperactive or nonhyperactive. A <
rating in the positive direction was accepted only if the -i
parent contended that the subject's excessive activity level
was a primary problem area. This information was gathered
in the follow-up debriefing conference.
Secondly, the three groups were compared on each of the
•three factors. One-way analysis of variance was performed
for each factor across the three groups with the following
results: Excitability, F (2,68) = 45.76, p<.0001; Motor
Coordination, F (2,68) = 5.37, p<.006; and Directed Attention
F (2,68) = 37.72, p<.0001. Thus, the overall F-statistic is
highly significant for all three factors. The specific pair-
wise comparisons are summarized in Table 3. In all cases,
the separation was greatest between the hyperactive group
and the control group with the nonhyperactive clinical group
occupying the middle position. Mothers of hyperactive chil
dren attributed significantly more hyperkinetic characteris
tics to their children than did the mothers of the subjects
in the other two groups.
mmmmm
58
TABLE 3.—Pairwise Comparisons of Hyperactive (H), Nonhyperactive (NH), and Control (C) Groups on Three Zukow Rating Scale Factors
Factor
Excitability
H vs NH
H vs C
NH vs C
Motor Coordination
H vs NH
H vs C
NH vs C
Directed Attention
H vs NH
H vs C
NH vs C
df
40 5.13 0001
51
51
40
51
51
40
51
51
9.56
3.90
1.40
3.24
1.69
2.47
8.31
5.58
.0001
.0002
ns
.001
ns
.02
.0001
.0001
•I
I
Measure of Intelligence
The Slossen Intelligence Test (SIT) is a brief individ
ual test of intelligence. Items are adapted from the
Stanford-Binet. The test takes approximately 20 minutes to
administer. All items are presented verbally and require
spoken responses. The mean IQ score on the SIT is 100, with
a standard deviation of 25. Hunt (1972) reported a test-
retest reliability of .97 within a two-month interval. The
validity coefficient in relation to the Stanford-Binet Form
L-M is .92 for ages 4-19. Item content stresses mathematical
reasoning, vocabulary, auditory memory, and information. Of
59
the 71 subjects, 62 were administered the Slossen Intelli
gence Test. The remaining 9 subjects were given the
Wechsler Intelligence Scale for Children-Revised.
Procedure
Subjects in the two clinical groups had been referred
for psychological evaluation to the Psychology Clinic on the <
Texas Tech University campus. Research data was gathered
during this "routine" evaluation providing mothers of the
children agreed to participate in the research project along
with their child by signing a written consent form (Appendix
E). Mothers of subjects participated by completing two mea
sures, the Behavioral Style Questionnaire and the Zukow
Rating Scale of Hyperkinesis. Although both instruments are
self-administered, the instructions were briefly summarized
and the rater was asked if she had any questions. Questions
pertaining to the nature of the inventories were deferred
until the questionnaires had been completed to insure stan
dardized administration of the tests and to avoid possible
biasing effects. While the parent was completing these
inventories, the child was administered an individual intel
ligence test and the Vigilance Task.
Mothers of control group subjects were asked to partic
ipate in the research project via a letter sent home from
school with their child (Appendix A). The letter accom
panied a packet of materials. Mothers electing to
60
participate completed and returned (a) the BSQ accompanied
by the standard instructions, (b) the Zukow Rating Scale,
(c) a short data sheet to be completed on the child, includ
ing name, birthdate, sex, grade in school, and clinical
status (i.e., whether or not the child was currently being
treated for emotional or behavioral problems), and (d) a
1'
signed informed consent form. « • " »
Children in the control group were administered the SIT -j I
and Vigilance Task. Testing was carried out either in a "
mobile laboratory owned by the Department of Psychology
which was parked on the premises of one of the elementary
schools or in the Psychology Clinic on the Texas Tech Univer
sity campus. The same apparatus and instruments were used
in both settings.
The entire data-gathering process took approximately
one and one-half hours per subject. The results were shared
with parents of each subject after the evaluation was com
pleted. This follow-up contact also served as the study's
debriefing phase with the aims of the research project being
discussed with the parents.
Statistical Hypotheses
The intent of the present study was to empirically in
vestigate the relationship between tem.perament and behavioral
disorders in children, especially hyperactivity.
61
The statistical hypotheses were as follows:
1. The Difficult Child pattern (arhythmicity, withdrawal,
slow adaptability, intense reactivity, negative mood)
will occur with significantly higher probability in the
hyperactive and nonhyperactive clinical groups compared
to the control group. Furthermore the Difficult Child
pattern will be a significant predictor of membership J
in the clinical groups compared to the control group. j
2. The Active Difficult Child pattern (high activity, l a 9
arhythmicity, withdrawal, slow adaptability, intense J
reactivity, negative mood) will occur with significantly
higher probability in the hyperactive group compared to
the nonhyperactive clinical group and the control group.
Furthermore, the Active Difficult Child pattern will be
a significant predictor of membership in the hyper
active group compared to the nonhyperactive groups.
CHAPTER III
RESULTS
/
The research hypotheses pertain to the efficacy of
specific trait patterns to predict group membership. This
is primarily a classification problem. Therefore, discrim
inant analysis was deemed the most appropriate statistical %
tool for analyzing the data. The BMDP7M computer package
program for stepwise discriminant analysis (Dixon, 1977)
was selected for this purpose. Briefly, discriminant anal
ysis is a statistical procedure for classifying individuals
into groups on the basis of their scores on the predictor
variables. The analysis generates a discriminant function
which gives the "best" prediction of the "correct" group
membership of each member of the sample (Kerlinger, 1973).
The findings for the study's two hypotheses are pre
sented separately with greater emphasis given to the main
hypothesis.
Hypothesis 1
The first hypothesis concerned the predicted relation
ship between the five variable Difficult Child temperament
pattern and the presence of behavioral disorders in chil
dren (Thomas, Chess & Birch, 1968). Since no distinction
was made regarding type of disorder, the two clinical groups
were combined and the resulting criterion group was
62
m^.
63
designated "clinical" (N=40). The second criterion group
was the control (N=31). Seventy-one subjects comprised the
total sample.
The five predictor variables in the Difficult Child
pattern were Rhythmicity, Approach/withdrawal, Adaptability,
Intensity, and Mood. All variables were continuous, having
the same range of values from one (not difficult) to six «
(very difficult). The means and standard deviations for i
each variable for the clinical and control groups, as well
as for the total sample, are provided in Table 4. The I 1
clinical group means were higher on all five variables. The
correlation matrix for the six temperament variables consid
ered in the present investigation are presented in Table 5.
Can these five variables, taken one at a time, discrim
inate between the clinical and control groups? Ryan (1972)
has recommended that when a pattern of variables based upon
theory is tested, the experimentwise error rate is most
appropriate to reduce Type I errors. Five variables comprise
the Difficult Child pattern and therefore, alpha level v/as
set at .01. Two variables. Adaptability and Mood, reached
highly significant levels. Rhythmicity and Approach/with
drawal approached significance, while Intensity was clearly
a poor discriminator between these two groups. The results
of the univariate analyses are summarized in Table 6.
64
TABLE 4 .—Means and S t a n d a r d D e v i a t i o n s f o r F i v e V a r i a b l e s i n D i - f f i c u l t P a t t e r n f o r C l i n i c a l and C o n t r o l Groups
Variable
Clinical
3.07 .66
3.02 1.02
3.43 .89
4.57 .66
3,79 .79
Group
Control
2.68 .57
2.52 .79
2.25 .75
4.45 .59
3.01 .74
Total
2.90 .62
2.80 .93
2.92 .83
4.52 .63
3.45 .77
R h y t h m i c i t y Mean S .D.
H
Approach f-Mean S.D. 1.02 .79 .93 I
Adaptability Mean S.D.
Intensity Nean S.D.
Mood Mean S.D.
TABLE 5 . — C o r r e l a t i o n M a t r i x f o r Temperament V a r i a b l e s (N=71)
A c t . Rhythm. Approach A d a p t . I n t e n s i t y Mood
A c t i v i t y . 1 9 - . 0 6 .30 - . 1 4 . 44
Rhythm, -13 .26 . 0 8 . 3 8
A p p r o a c h -35 - . 0 3 .17
A d a p t . . 2 1 . 7 2
I n t e n s . -29
Mood
S i ' ;
6 --J
TABLE 6.—Summary of One-way ANOVA R e s u l t s Between C l i n i c a l and Con t ro l Groups on D i f f i c u l t Chi ld V a r i a b l e s
Name of Var iab le df
Rhythmicity
Approach
Adaptability
Intensity
Mood
1.69
1.69
1.69
1,69
1.69
6.828
5.029
34.790
0.666
17.648
.02
.03
.001
ns
.001
The atypical nature of the control group required that
three non-temperament variables be included in the statis
tical analysis. When compared to the clinical subjects,
the control subjects obtained higher intelligence test
scores, came from homes reflecting a higher socioeconomic
strata, and were more likely to be residing with both biolog
ical parents (Table 7). All three of these extraneous vari
ables were found to differentiate between the clinical and
control groups at statistically significant levels (Table 8)
In order to determine the actual contribution of the
Difficult Child pattern to the discrimination between the two
groups, two separate and sequential analyses were performed.
First, could the five temperament variables together with
the three extraneous subject variables discriminate between
clinical and control group subjects? A full Model discrim
inant analysis was performed which included all eight vari
ables. The Wilks' Lambda criterion for testing the
c y
66
significance of the overall difference among the group cen-
troids, indicated the equally of mean vectors for the
clinical and control groups was untenable (0.529). When
this statistic was transformed into an approximate F-statis
tic, it yielded an F (8,62) = 6.87, which was significant
at the .001 level, indicating the eight variables taken
together statistically discriminated between the two groups.
TABLE 7,—Means and Standard Deviations for Three Extraneous Variables for the Clinical and Control Groups
Variable Group Clinical Control Total
IQ Mean 103.70 S.D, 14.36 a
SES Mean 3.90 S.D, 1.54
b Family Status Mean .42 S.D. .50
20 12
2
. 4 1
. 6 3
. 5 4
.80
. 09
. 30
1 1 1 . 13 .
3. 1.
.00
.63
.30
.28
.28
.42
^Socioeconomic Status (SES) was estim.ated from parent occupation which was ranked on the modified Hollingshead's Two Factor Index of Social Position (Myers & Bean, 1968), with a range of values from one (high prestige) to seven (low prestige).
^Family status was coded either 0 (intact) or 1 (non-intact). Therefore, the means can be read as percentages of biologically non-intact families.
V «• V
>
1 f
» » J «
67
TABLE 8.—Summary of One-way ANOVA R e s u l t s Betv;een C l i n i c a l and C o n t r o l Groups on Three E x t r a n e o u s S u b j e c t V a r i a b l e s
Name of V a r i a b l e df
IQ
SES
Family Status
.69
.69
.69
26.260
19.434
10.399
.001
.001
.001
2 tor variables and the criterion groups was .68. The R
value (.47) indicated that the Full Model (eight variables)
accounted for 47% of the separation between the clinical
(0.817) and control (-1.054) group centroids.
The second part of the analysis determined the unique
contribution of the five variable Difficult Child pattern
to the discrimination between the two groups. The three
extraneous variables (Reduced Model) alone accounted for 2
38% of the separation between the two groups (R =.38). The
difference in the squared multiple correlations between the
9 2 Full Model (R =.47) and the Reduced Model (R =.38) was the
unique contribution to the discrimination due to the five 2
variable temperament pattern. The resulting R for the
Partial Model (five temperament variables) was .09. Finally,
with the contributions of IQ, parent occupational level, and
family status partialed out, the following formula yields a
multivariate F for the Difficult Child pattern:
w
>
The multiple correlation (R) between the eight predic- i •
u >
I ) I
68
2 p _ Rpartial/df M u l t i v a r i a t e " 'ZTTTTy T T (Kerlinger & Pedhazur,
R full/N-k-1 1973)
The resulting F (5,62) = 2.15, £<.08, was in the borderline
area of statistical significance. The partial R^ (.09)
indicated that only 9% of the separation between the clin
ical and control groups could be accounted for by the Diffi- ^ c
cult Child pattern.
To summarize the findings for the first hypothesis, the
Difficult Child pattern alone did not differentiate at a
statistically significant level, between the clinical and
control groups. However, there was a trend in the hypoth
esized direction.
Hypothesis 2
The study's main hypothesis concerned the predicted
relationship between the six variable Active Difficult Child
pattern and a specific childhood behavioral disorder, hyper
activity. The total sample (N=71) was divided into two sub
groups of hyperactive and nonhyperactive subjects. In this
analysis the nonhyperactive clinical subjects were combined
with the controls. The resulting two criterion groups were
designated "hyperactive" (N=20) and "nonhyperactive" (N=51).
The six predictor variables in the Active Difficult
constellation were Activity level, Rhythmicity, Approach/
withdrawal, Adaptability, Intensity, and Mood. The higher
mm
69
the scores, the more pronounced the temperament pattern.
The means and standard deviations for each variable for both
the hyperactive and nonhyperactive groups, and for the total
sample, are presented in Table 9. As a group, hyperactive
children obtained higher scores on all variables in the pat
tern consistent with the hypothesized direction. ti
The results of the univariate analyses are summarized J k
in Table 10. Experimentwise error rate was utilized. Six |
variables constituted the Active Difficult Child pattern and
therefore, alpha was set at .008. Activity, Adaptability,
and Mood were all statistically significant. Of the remain
ing three variables, Rhythmicity closely approached signif
icance, while Intensity and Approach/withdrawal clearly did
not differentiate between groups.
The three extraneous variables--IQ, parent occupation,
family status—were again taken into account. The means and
standard deviations for each variable for both hyperactive
and nonhyperactive groups, and for the total sample, are
presented in Table 11. As a group, nonhyperactive children
obtained higher intelligence test scores, were from a higher
socioeconomic strata, and were more likely to be living with
both biological parents, than hyperactive subjects.
Univariate analyses were performed on these three vari
ables (Table 12). Subject's measured intelligence and parent
70
occupation were statistically significant, v/hile family
status was not a discriminating variable.
TABLE 9.—Means and Standard Deviations for Six Variables in Active Difficult Pattern for the Hyperactive and Nonhyperactive Groups
Variable Hyperactive
Group Nonhyperactive Total
Activity Mean S.D.
Rhythmicity Mean S.D.
Approach Mean S.D.
Adaptability Mean S.D.
Intensity Mean S.D.
Mood Mean S.D.
5.08 .62
3.20 .75
3.04 1.23
3.78 .73
4.71 .46
3.97 .78
3.62 .60
2.79 .57
2.71 .82
2,58 .90
4.44 .68
3.25 .81
4.03 .61
2.90 .62
2.80 .95
2.92 .86
4.52 .62
3.45 .80
TABLE 10.—Summary of One-way ANOVA Results Between Hyperactive and Nonhyperactive Groups on Active Difficult Variables
Name of Variable
Activity
Rhythmicity
Approach
Adaptability
Intensity
Mood
df
1.69
1.69
1.69
1.69
1.69
1.69
1
82.043
6.303
1.684
27.753
2.714
11.480
P
,001
.02
ns
.001
ns
.001
71
TABLE 11.—Means and Standard Deviations for Three Extraneous Subject Variables for the Hyperactive and Nonhyperactive Groups
Variable Group
IQ Mean S.D.
SES Mean S.D.
Family Status Mean S.D.
eractive
96.79 13.26
4.25 1.16
.40
.50
Nonhyperactive
116.56 13.23
2.94 1.37
.23
.42
Total
111.00 13.24
3.30 1.32
.28
.45
TABLE 12.—Summary of One-way ANOVA Results Between Hyperactive and Nonhyperactive Groups on Three Extraneous Subject Variables
Name of Variable df F p_
IQ
SES
Family Status
1,69
1,69
1,69
32,027
14.083
1.924
.001
.001
ns
A Full Model discriminant analysis was performed which
included the six temperament variables and the two signif
icant extraneous variables. The analysis yielded a Wilks'
Lambda criterion (0.375) which was transformed into an
approximate F-statistic. The resulting F (8,62) = 12.89,
was highly significant (p<.001).
72
The multiple correlation (R) between the eight predictor
variables and criterion groups was .79, yielding an R =.62. 2
The R value (.62) indicated that the Active Difficult Child
temperament pattern together with the two subject variables
(IQ, parent occupation) accounted for 62% of the separation
between the hyperactive (2.030) and nonhyperactive (-0.796)
group centroids.
A subsequent analysis was conducted to ascertain the
actual contribution of the six variable Active Difficult
pattern to the discrimination between hyperactive and non
hyperactive groups. The two extraneous variables (Reduced
2 Model) alone accounted for 34% of the group separation (R =
.34). The difference in the squared multiple correlations
2 2
between the Full Model (R =.62) and the Reduced Model (R =
.34) was the unique contribution to the discrimination which
could be ascribed to the Active Difficult temperament con-2
stellation. The Partial Model R (.28) was then tested for statistical significance. The resulting F (6,62) = 7.67,
2 was significant at the .001 level. The partial R indicated
that 28% of the separation between hyperactive and nonhyper
active groups was due to the six variables in the Active
Difficult pattern.
Taking into account all eight variables in the Full
Model, what is the relative contribution of each variable to
the discrimination between the two groups? VJhen the variables
mm
13
have different standard deviations, this is determined by
multiplying each weight by the standard deviation of the
corresponding variable (Bennett & Bowers, 1976). The result
ing products can then be compared for their relative contri
bution to the discriminant function. The discriminant
weights, the total sample standard deviations, and the
product of these two values is presented in Table 13. The
variables are listed in order of relative magnitude. The
discriminant function constant was -3.077. High activity,
slowness to adapt, intense reactivity, and relatively low
socioeconomic status, characterized the hyperactive children.
The nonhyperactive children were relatively brighter, more
regular in eating and sleeping habits, more positive in mood,
and more likely to approach unhesitantly a new situation.
TABLE 13.—Discriminant Weights, Standard Deviations, and Products, for the Eight Variables Discriminating Hyperactive and Nonhyperactive Groups
Name of Variable Discriminant Weight
X Standard Deviation
Produc-
Activity
IQ
Adaptability
Rhythmicity
Intensity
SES
Approach
Mood
1.24
-.03
.34
-.19
.17
.05
-.02
-.02
.61
13.24
.86
.62
.62
1.32
.95
.80
.76
-.40
.29
-.11
.10
.07
-.02
-.01
Variables are listed in order of descending magnitude of contribution to the discrimination between hyperactive and nonhyperactive groups.
74
How good a predictor was the discriminant function
which was based on the eight variable Full Model? The clas
sification matrix is presented in Table 14. Each subject
has been classified into either the hyperactive or nonhyper
active group. Shown in the Table are the number classified
into each group and the percentage of correct classifica
tions. Clearly, this eight variable model cojistituted a
highly discriminating set of predictor variables. In the
total sample, 91.5% of the children were correctly classified
on the basis of this pattern. The percentage of nonhyper
active children correctly classifed (92.2%) was only slightly
better than for the hyperactive group (90.0%).
How good would this discriminant function be when
applied to a new sample? An often raised concern (Huberty,
1975) is that the discriminant function weights derived from
one sample decline in their discriminating power when applied
to a new samLple. Several methods of estimating error rates
for a given discriminant function have been suggested
(Lachenbruch & Mickey, 1968) . Lachenbruch (1967) has pro
posed a technique known as the "jackknife" for estimating
the proportion of classification errors. This procedure in
volves successively omitting one case from the computation
of the discriminant function and subsequently classifying
that case according to the function thus derived.
75
TABLE 14.—Classification Matrix
Group Percent Correct
Number of Cases Classified into each Group
Hyperactive
(N=20)
Nonhyperactive
(N=51)
Total
(N=71)
90.0
92,2
91.5
Hyperactive
18
22
Nonhyperactive
2
47
49
The results of the jackknife validation procedure are
shown in the classification matrix in Table 15. Both the
hyperactive and nonhyperactive children were correctly
classified in 90% of the cases. This validation method
results in less than 2% reduction in correct classifications
for the total group.
A second method of estimating the error rate for a given
discriminant function without cross-validation is to calcu
late the "shrinkage" of the multiple correlation. When a set
of discriminant weights derived in one sample are applied to
the predictor scores of another sample the resulting R will
almost always be smaller than the original R. If the present
set of discriminant weights were applied to a new sample, the 2
following formula provides the estimated shrinkage m the R
value:
76
H = l-(l-R^)(^§^) (Kerlinger & Pedhazur, 1973)
TABLE 15.—Jackknifed Classification Matrix
Group
Hyperactive
(N=20)
Nonhyperactive
(N=51)
Total
(N=71)
Percent Correct
90.0
90.2
90.1
Number of Cases Classified into each Group
Hyperactive Nonhyperative
18
23
46
48
The estimated shrinkage of the squared multiple corre-
lation (R ) between the Full Model (eight predictors) and
the hyperactive and nonhyperactive groups was .57. The dif-
2 ^2
ference between R (.62) and R (.57) showed a decrease of
.05, indicating that the set of discriminant weights derived
from the present sample are highly stable, and therefore,
generalizable to other similar samples with only negligible
loss of predictive power.
To summarize the findings for the second hypothesis,
the Active Difficult pattern alone significantly discrim
inated between hyperactive and nonhyperactive groups of chil
dren. The six temperament variables were then combined with
the two significant non-temperament factors, IQ and parent
»«IIH
77
occupational level. A discriminant analysis based on these
eight predictor variables resulted in better than nine out
of ten subjects being correctly classified into their
appropriate group.
II
CHAPTER IV
DISCUSSION
The purpose of this study was to investigate the rela
tionship between two specific temperament patterns and the
presence of behavioral disorders in children. Two separate
hypotheses were tested. The first hypothesis derived from
the theoretical formulations of Thomas, Chess and Birch
(1968, 1970). These investigators have asserted that a
specific constellation of temperament characteristics,
designated the Difficult child, was associated with a high
incidence of behavioral pathology in Children. The second
hypothesis investigated in the present study was that the
temperament pattern of the highly active Difficult child
would correspond to clinical hyperactivity. While the
relationship between the active Difficult child and the hyper
active child had been conjectured (Goggin, 197 2; McDevitt &
Carey, 1978), heretofore, this relationship had not been
empirically investigated.
Hypothesis 1
The hypothesis that children with behavioral disorders
are likely to be characterized as Difficult children did not
receive confirmation. As a multivariate pattern, the Dif
ficult child constellation failed to discriminate between
78
79
clinical and "normal" control group subjects at a statis
tically significant level.
Do these five variables, taken one at a time, differen
tiate between clinical and control subjects? Univariate
analyses revealed that four of the five temperament charac
teristics comprising this pattern either exceeded or closely
approached statistically significant levels. Group mean
scores on all Difficult child categories were higher for the
clinical group consistent with the hypothesized direction.
Children exhibiting clinical problems tended to be perceived
by their mothers as slow to adapt to environmental demands,
predominantly negative in mood, biologically arhythmical and
inclined to withdraw from new experiences. The results of
the univariate analyses revealed that, of the five tempera
ment traits in the Difficult pattern, only Intensity failed
to differentiate clinical subjects from controls.
Further univariate analyses demonstrated that three non-
temperament factors also differentiated clinical and control
subjects. These factors were IQ, parent occupational level,
and intactness of family of origin. As a group, control
children obtained significantly higher intelligence test
scores, their parents held jobs reflecting a higher socio-
econom.ic position, and the subjects were more likely to be
living with both biological parents.
impwip
80
These three non-temperament factors combined with the
five temperament variables comprising the Difficult pattern
clearly discriminated clinical from nonclinical subjects.
However, when the effects of the three non-temperament vari
ables were statistically partialed out, the unique contribu
tion of the five temperament variables was markedly
diminished. The Difficult pattern did not differentiate
between clinical and control group subjects at a statistically
significant level. The significant univariate results for
the temperament variables were, therefore, confounded by the
effects of the three extraneous variables. The actual dif
ferences between the two groups accounted for by the Difficult
pattern is only suggestive in light of the statistical trend
in the hypothesized direction.
Hypothesis 2
The hypothesis that children who are clinically hyper
active are likely to be characterized as highly active
Difficult children has received qualified support. This six
variable temperament pattern proved to be a highly signif
icant discriminator between hyperactive and nonhyperactive
subjects. However, univariate analyses revealed that only
three of the six temperament characteristics actually dif
ferentiated at statistically significant levels. It is
hardly surprising that the hyperactive group was perceived
81
by their mothers as extremely active. More meaningful is
the finding that this group was characterized as much less
adaptable and more likely to be moody, fussy, and unfriendly
than the nonhyperactive group. There was also a tendency
to attribute greater irregularity in sleeping and eating
habits to the hyperactive group. The two remaining temper
ament traits. Intensity and Approach/withdrawal, proved to
be poor discriminators between the two groups.
Two non-temperament factors—IQ and parent occupational
level--also differentiated between hyperactive and nonhyper
active subjects at statistically significant levels. As a
group, nonhyperactive children earned significantly higher
intelligence test scores and the occupations of their parents
reflected a higher socioeconomic strata.
When the six variable Active Difficult child temperament
pattern was combined with the two non-temperament factors,
the resulting eight variables clearly differentiated hyper
active from nonhyperactive subjects. When the effects of
the two non-temperament variables were statistically par
tialed out, the unique contribution of the Active Difficult
Child variables remained highly significant.
A prediction equation consisting of these eight vari
ables correctly classified better than nine out of ten chil
dren into their respective groups. Both hyperactive and
nonhyperactive subjects were classified with nearly equal
82
accuracy (90-92%). Moreover, the prediction equation was
highly stable so bhat when it is applied to a new sample
with similar characteristics, there should be only negligi
ble loss of predictive power.
Theoretical Implications
The investigator was primarily interested in the hypoth
esized relationship between hyperactivity and the Active
Difficult child temperament pattern. The results suggest the
diagnostic category hyperactivity may be reformulated in
terms of the Thomas et al. (1968, 1970) theory of child devel
opment. A major advantage of recasting the hyperactive syn
drome in term.s of behavioral style is that the temperament
characteristics of the active Difficult child can be under
stood within the framework of a general theory of child
development.
A fundamental tenet of temperament theory is that the
adequacy of the child's functioning is dependent upon the
degree the expectations, demands and opportunities in the
environment are in accord with the child's own characteris
tics, including behavioral style, needs, abilities, and
unique learning history. Thomas and Chess (1977) refer to
this principle as "goodness of fit." A "good fit" exists
when there is congruence between environmental press and the
child's capacity to cope constructively with these demands.
83
When there is a "poor fit," dissonance results and the child
experiences stress. This is a homeodynamic principle in
which development is viewed as a process of continual change
and expanding competency. Dissonance and stress are inte
gral facets of the normal growth process. On the other hand,
prolonged or extreme dissonance may lead to emotional and
behavioral disorders in the developing child.
The present findings can be meaningfully interpreted
within this theoretical framework. The hyperactive group of
children were perceived as possessing several of the Active
Difficult child characteristics. They were characterized as
slow to accommodate to environmental demands, predominantly
negative in mood, extremely active, and inclined to exhibit
irregular eating and sleeping habits. Together, these four
temperament characteristics constitute a style of inter
acting with the environment. In the remainder of this sec
tion, each of these temperament characteristics is elaborated
upon within the context of the "goodness of fit" principle of
adaptive functioning.
Adaptation is defined by Thomas et al. (1968) as the
child's capacity to respond readily to environmental demands
to change.* The following items from the Behavioral Style
*Adaptability is used here in the restrictive sense of accommodation. In a broader evolutionary sense, adaptability implies both the organism accommodating to the environment as well as affecting changes in the environment to satisfy needs.
II II IJ
84
Questionnaire are representative of this temperament cate
gory: "The child needs a period of adjustment to get used
to changes in school or at home"; "The child seems to take
setbacks in stride"; "The child will avoid misbehavior if
punished firmly once or twice." The adaptable child, in
this sense, demonstrates flexibility and a capacity to
respond promptly to environmental pressures to accommodate.
Hyperactive children, on the other hand, are perceived as
inflexible and unaccommodating. Thus, situations in which
these children are pressured to make rapid adjustments are
likely to lead to conflict and stress. All of the children
in the present study were in the process of adjusting to
school. Elementary school is a major developmental challenge
even for the adequately functioning child. There are pres
sures to adapt to a new physical setting and to strange
adults in unfamiliar roles, to master increasingly complex
cognitive tasks, to accommodate to a host of new rules and
regulations, and to adjust to the socialization pressures
of a large group of unfamiliar peers (Thomas & Chess, 1977).
For the child who characteristically adjusts only slowly to
changes, the early school experience may be fraught with
stress.
For the energetic, highly active youngster the class
room may pose a special kind of challenge to the child's
coping capacities. Spontaneous motor expression is severely
mmm
85
restricted in the traditional classroom setting. Environ
mental demands to sit still, pay attention, and complete
assignments in one sitting, may be experienced by the child
as impossible demands. Furthermore, the child's lack of
compliance may be interpreted by teachers as indications of
willful disobedience and lead to frequent scoldings and
punishment. Many of the school children in the study's
hyperactive group, even though only at the beginning of
their school experience, were already singled out as manage
ment problems by their teachers.
The finding that hyperactive children were perceived
as irregular in both eating and sleeping habits again sug
gests a discrepancy between the child's behavioral style
and environmental demands. The arhythmical child is out
step with his/her environment and is likely to experience
stressful interactions in the most fundamental areas of
daily living—eating and sleeping.
The hyperactive children were also perceived as moody,
fussy, and negativistic. The following items from the
Behavioral Style Questionnaire are representative of the
Mood category: "The child is moody for more than a few
minutes when corrected or disciplined"; "The child laughs
or smiles while playing"; "The child cries or whines when
frustrated." In terms of the "goodness of fit" model, the
child whose affect is predominantly unpleasant to others
86
stands a greater chance of eliciting critical and disapprov
ing responses from the environment. Again, this increases
the child's vulnerability to stress.
In the preceding paragraphs each of the temperament
characteristics differentiating the hyperactive from the non
hyperactive group has been examined separately. Possible
stress patterns have been outlined. In fact, these charac
teristics constitute a behavioral pattern. The child per
ceived as highly active, unaccommodating, moody, and
irregular in eating and sleeping habits, is "at risk" to
develop emotional and behavioral disorders. The children in
the present study have already been identified as problem
children. However, it is important to recall that tempera
ment organization "modifies the degree of risk, but does
not directly convert risk to reality" (Thomas et al., 1968).
The critical ingredient is the interaction between the child
with a given behavioral style and the environment. For exam
ple, one parent may respond to the active Difficult child
with patience, firmness and consistency. This diminishes
the likelihood that a pattern of stress will develop, A
second parent may react angrily to the child's fussiness,
insist upon immediate compliance, and so on. Such a child
has a risk of developing psychological problems.
Three non-temperament factors were identified in the
present study that correlated with the presence of behavioral
mmmm
87
disorders in children. Disturbed children were more likely
to obtain lower intelligence test scores, come from a lower
socioeconomic strata, and to have experienced greater family
instability. Although the interaction between the child's
temperament and these non-temperament factors was not
directly investigated, a reasonable inference is that the
child exhibiting Difficult characteristics will be more "at
risk" in an environment characterized by greater financial
uncertainties and family instability.
The interaction between the temperament characteristic,
high activity level, and the environment, was illustrated by
Thomas and Chess (197 7) . These investigators observed that
only a single child in their affluent suburban New York
sample exhibited excessive and uncontrolled motor activity.
In contrast, in their urban Puerto Rican working class sample,
53% of the children who developed behavioral problems pre
sented as "hyperactive." Importantly, there was no differ
ence in the relative incidence of Difficult children in the
two samples. Thomas and Chess (1977) explained this finding
by pointing to the environmental differences between these
two groups. Children in the former group were raised in an
environment characterized by ample indoor and outdoor living
space. On the other hand, the Puerto Rican children had to
accommodate to the extremely restricted living space of
crowded urban apartments. For energitic children, this
88
condition constituted a severe stress. The child with a
highly active behavioral style was at odds with the demands
and expectations of the environment. According to Thom.as
and Chess, the resulting dissonance would account for the
difference in incidence of "hyperactivity" between these
two groups of children.
The "goodness of fit" conceptualization of the inter
action between hyperactive child and environment also re
sembles the ecological model proposed by Williamson et al.
(1980). Both models place an emphasis upon reciprocal inter
action. The ecological model describes this interaction in
terms of a continuous feedback loop between child and envi
ronment. When a negative feedback loop develops, the child's
sense of worth and self-esteem may suffer, setting the stage
for behavioral and emotional problems secondary to the hyper
activity. Likewise, the emergence of a positive feedback
loop signifies an environmental acceptance of the child's
behavioral style, increases the consonance between child and
world, and reduces the risk of psychopathology. Thus, the
present findings also provide support for the ecological
model of hyperactivity.
To summarize this section, hyperactivity may, in part,
be redefined as a characteristic behavioral style. This
allows us to proceed beyond a simple listing of the "sym.p-
tomatic" behaviors to a broader understanding of the child's
89
characteristic style of interacting with his or her environ
ment. The finding that two non-temperament factors— mea-'
sured intelligence and parent occupational level—also
distinguished hyperactive from nonhyperactive children,
underscores the need to evaluate the child's behavioral style
in relationship to other person (e.g., IQ) and environmental
(e.g., SES) influences. Adequacy of functioning depends
upon the interaction between temperament, intellective fac
tors, and other characteristics of a particular child, and
significant factors in the child's environment. Thus,
describing the clinically hyperactive child in terms of the
temperament characteristics of adaptability, mood, activity
level, and rhythmicity, may advance our understanding of
this prevalent childhood disorder.
Clinical Implications
Reformulating clinical hyperactivity as a behavioral
style may be advantageous to the clinician as well as to the
child. The child who is diagnosed as hyperactive, whether
by the mental health professional or the classroom teacher,
is typically perceived as "containing" the disorder. The
interventions that follow are usually aimed at changing the
child's behavior. The therapy, whether it involves drug
treatment or behavior modification, is typically child-
centered.
90
Reframing hyperactivity in terms of behavioral style
emphasizes the interaction between the child and his or her
environment, and offers the clinician a means of pinpointing
areas of potential stress. For example, the hyperactive
children in the present study were likely to experience
stress due their relatively slow adaptability and frequent
negative affective expressions. These temperament qualities
tend to draw disapproving and critical responses from parents,
teachers and classmates. This, in turn, increases the likeli
hood of the child responding with further negative affect and
increased resistence to change. The resulting stress cycle
may eventuate in the child developing long-standing behavioral
and emotional problems. Children diagnosed as hyperactive
are prone to continuing adjustment problems in later years
(Laufer, 1971). Focusing on the interaction between the
child's behavioral style and the expectations and demands in
the environment, therapeutic efforts would be directed at
lessening this dissonance. Intervention would emphasize
modification of the patterns of interaction between child
and environment. Chess, Thomas, and Birch (1965) have
offered detailed advice to parents on how to raise children
with different behavioral styles.
The results of the present investigation also have im
plications for a prevention model of mental health. The
study results suggest a means for the early identification
II II II
91
of a high risk population during the preschool or early
schodl years. Although the present study did not address
the longitudinal stability of the Active Difficult pattern,
Carey and McDevitt's (1978) findings suggest that this par
ticular temperament organization is more persistent and
less amenable than other patterns to the usual developmental
modifications that occur as the child grows up. Carey and
McDevitt found that a disproportionate number of individuals
who were Difficult as 4- to 8-month old infants remained so
throughout the period of 3 to 7 years. Furthermore, they
found that the two temperament characteristics that best pre
dicted stability of this pattern over time were high activity
(p<.01) and very negative mood (p<.01). These findings give
support to the position that early identification of children
with the Active Difficult temperament organization may be a
desirable goal if it facilitates early intervention and
lessens the likelihood of subsequent psychological
disturbance.
Practical Application
The Behavioral Style Questionnaire (BSQ) was demonstrated
to be an effective instrument for differentiating hyperactive
youngsters from a mixed group of "normal" children and chil
dren with nonhyperactive behavioral problems. The discrim.-
inant analysis produced a set of weights which maximized
the separation between these two groups. The equation
92
consisted of the six variables in the Active Difficult pat
tern and two non-temperament variables, a measure of the
child's intelligence and an estimate of the child's socio
economic status. Each of the eight variables was assigned
a weight: Activity (1.24), IQ (-.03), Adaptability (.34),
Rhythmicity (-.19), Intensity (.17), SES (.05), Approach/
withdrawal (-.02), and Mood (-.02).
How the clinician or researcher can utilize these weights
in conjunction with the individual's BSQ scores is detailed in
the paragraphs that follow.
The following formula reduces the six variables compris
ing the Active Difficult pattern and the two non-temperament
variables to a single composite score:
Y = a + w^(X^) + W2(X2) + ^^^^3^ * * * " 8 ^8^'
where "Y" is the composite score,
where "a" is the equation constant, -3.07,
where w,...w^ are the discriminant weights for the 1 o
eight variables,
where X^ ...X^ are the individual's raw scores for the 1 o
eight variables in the equation.
Once an individual's composite score is computed, the
next step is to determine group assignment—hyperactive or
nonhyperactive. Bennett and Bowers (1976) have suggested
that when the risks of misclassification are the same for
««pp«
93
both groups, than the optimum cut-off point is midway be
tween the means of the two groups. In the present case, the
mean of the hyperactive group was +2.03 and the nonhyper
active group mean was -0.79. This locates the cut-off point
at +0.62.
The individual's composite score can now be located on
a continuum of values with +.62 as the midpoint between the
group means. Using percentiles of the standardized normal
distribution (the total distribution of scores have been
standardized with a mean equal to zero and the standard
deviation equal to one), 90% of hyperactive children will
have composite scores ranging in value from +.75 to +3.31.
For nonhyperactive children, 90% will earn scores between
-2.07 and +.49. Since this set of weights produced 91,5%
correct classifications, there is no overlap in the two
distributions at the 90th percentile level.
Thus, the results of this study have practical value
for both clinician and researcher. Utilizing the set of
weights, the group means, and the suggested cut-off value,
an individual's scores on the BSQ in conjunction with a
measure of intelligence and the estimated SES value, can be
used as a means of accurately determining hyperactive/
nonhyperactive status.
94
Future Research Directions
The major methodological drawback in the present study
was the atypical nature of the control group. Ideally, the
comparison group would have been a random sample of public
school children from home environments similar to the hyper
active subjects. Had the control subjects more closely
approximated the hyperactive subjects, the study's outcome
would have been strengthened. To some extent, the lack of
equivalency between these two groups was controlled for sta
tistically. Nevertheless, the mismatch between "normal" and
clinical subjects severely limits the generalizability of
the findings. The next research step should be the replica
tion of the present study using a more equivalent control
group of "normal" children. If the groups are similar with
regard to the non-temperament variables IQ and SES, then
these variables could be deleted from the prediction equation
presented in the last section. This adjustment would greatly
simplify the computations of the clinician or researcher who
wished to use the BSQ as a diagnostic instrument.
Intelligence, socioeconomic background, and family sta
bility, were found to be important non-temperament factors
in the present investigation. Future research efforts
exploring the relationships between temperament and psycho
pathology would profit by systematically studying the
95
relationship between behavioral style and other non-
temperament factors.
The results of the current investigation offer support
for the hypothesis that hyperactive children may be
described in terms of a behavioral style derived from tem
perament theory of child development. However, it is impor
tant to recognize that these findings were predicated on
the mother's perception of her child, rather than on the
child's observed behavior. Parent ratings may be subject
to bias and distortion. For example, the mother who is con
tinually at odds with her child may be prone to exaggerate
the child's maladaptive temperament characteristics. There
fore, future research efforts in this direction would be
strengthened by gathering independent behavioral samples in
addition to the questionnaire material. Another method of
enhancing the validity of the temperament description would
be the collection of independent ratings from several dif
ferent sources, including the child's mother, father and
classroom teacher.
The characteristics constituting the Active Difficult
temperament organization are identifiable in infancy (Carey
& McDevitt, 1978). Insofar as this temperament pattern
overlaps with clinical hyperactivity, determining the tem
perament profiles of preschool children may be a means of
alerting caregivers to an "at risk" population of children.
II
96
Longitudinal research is needed to determine the validity
of this developmental linkage. Is the Active Difficult
infant the precursor of the hyperactive school child? Fur
thermore, once these children are identified, will preven
tive intervention reduce the incidence of hyperactivity?
In conclusion, the present investigation was an ex
ploratory effort to establish a linkage between temperament
and a specific childhood disorder. The results offer support
for further investigation in this direction of linking behav
ioral style and psychopathology in children.
CHAPTER V
SUMMARY AND CONCLUSIONS
This study investigated the relationship between two
specific temperament patterns and the presence of behavioral
disorders in children. A literature review was presented
which included a survey of different etiological models of
hyperactivity and an explication of the Thomas, Chess, and
Birch (1968, 1970) temperament theory of child development.
Following from this review two separate hypotheses were
generated.
The first hypothesis derived directly from the Thomas
et al. theoretical formulation of child development. These
authors identified a specific cluster of temperament charac
teristics, designated the Difficult child, which they found
to be associated with a high incidence of behavioral path
ology. The validity of this relationship was examined. The
second hypothesis investigated was that the temperament
pattern of the highly active Difficult child corresponded to
clinical hyperactivity. The relationship between this spe
cific temperament pattern and hyperactivity had heretofore,
not been subjected to empirical test.
Seventy-one children and their mothers participated in
this study. An ex post facto research design was employed
97
98
which basically involved two independent steps: (a) the
assignment of each subject to the appropriate clinical
group, i.e., hyperactive or nonhyperactive, or the control
group, and (b) the assessment of the child's temperament by
means of the Behavioral Style Questionnaire (McDevitt &
Carey, 1978) . Other subject data included an assessment of
intelligence, a behavioral measure of attentional deficit
and motoric activity level, and a problem checklist.
Each hypothesis involved a two-group comparison. In
the first hypothesis all clinical subjects were compared to
the "normal" controls. For the second hypothesis the hyper
active group was compared to the composite group of nonhyper
active subjects. The predictor variables were the temperament
and non-temperament variables used to predict subject's group
membership.
Data from the two hypotheses were analyzed using step
wise discriminant analysis. The conclusions of this study
were as follows:
1. As a group, children with behavioral/emotional
problems tended to be perceived by their mothers
as Difficult children. However, this trend was
only suggestive.
2. A positive relationship was found between the
Active Difficult temperament pattern and children
who exhibit hyperactivity as a dominant sympton.
99
As a group, hyperactive subjects were perceived as
slow to adapt, moody, highly active, and irregular
in eating and sleeping habits.
3. Two non-temperament factors, intelligence and
socioeconomic status, also differentiated hyper
active from nonhyperactive children. This finding
underscores the need for a multideterminant model
of hyperactivity.
4. The results suggested hyperactivity may, in part,
be reformulated in terms of a characteristic behav
ioral style. This conceptual shift allows the
clinician to proceed beyond a simple listing of
symptomatic behaviors to a broader understanding
of the hyperactive child's characteristic style of
interacting with his or her environment.
5. A method for the identification of a high risk pop
ulation of children during the preschool and early
school years is offered. Prediction formula and
cut-off scores for the Behavioral Style Question
naire are provided.
REFERENCES
Anderson, R. P. A neuropsychogenic perspective on remediation of learning disabilities. Journal of Learning Disabilities, 1970, 3_/ 143-148.
Anderson, R. p. Hyperactivity. Unpublished manuscript, Texas Tech University, 1977.
Anderson, R. p., & Halcomb, C. (Eds.). Learning disability/ minimal brain dysfunction syndrome. Springfield: Charles C. Thomas, 197 6.
Anderson, R. P., Halcomb, C., & Doyle, R, The measurement of attentional deficits. Exceptional Child, 1973, 3_?.' 534-538.
Anderson, W. W. The hyperkinetic child: a neurological appraisal. Neurology, 1963, 13_, 968-973.
Bennett, S., & Bowers, D. An introduction to multivariate techniques for social and behavioral sciences. New York: John Wiley & Sons, Inc., 1976.
Cameron, J. R. Parental treatment, children's temperament, and the risk of childhood behavioral disorders. American Journal of Orthopsychiatry, 1978, 48(1), 140-147.
Campbell, S. B., Schleifer, M., & Weiss, G. Continuities in maternal reports and child behaviors overtime in hyperactive and comparison groups. Journal of Abnormal Child Psychology, 1978, 6_(1) , 33-45.
Cantwell, D- Genetics of hyperactivity. Journal of Child Psychology and Psychiatry, 1976, 16_, 261-264.
Carey, W. B. Clinical applications of infant temperament measurements. Journal of Pediatrics, 1972, 8_1, 823-828
Carey, W. B., & McDevitt, S. C. Stability and change in individual temperament diagnoses from infancy to early childhood. Journal of the American Academy of Child Psychiatry, 1978, 2 , 331-337.
Chess, S. Diagnosis and treatment of the hyperactive child. New York State Journal of Medicine, 1960, 6£, 2379-2385. •
100
101
Chess, S. Temperament and learning ability of school chil-^^®^- American Journal of Public Health, 1968, 58(12), 2231-22397 — —
Chess, S., & Hassibi, M. Behavior deviation in mentally retarded children. Journal of the American Academy of Child Psychiatry, 1970, 9_, 282-297.
Chess, S., & Hassibi, M. Principles and practice of child psychiatry. New York: Plenum Press, 1977.
Chess, S., Korn, S., & Fernandex, P. Psychiatric disorders of children with congenital rubella. New York: Brunner/Madzel, 1971.
Chess, S., Thomas, A., & Birch, H. G- Your child is a person. New York: Penguin Books, 1965.
Chess, S., Thomas, A., Rutter, M., & Birch, H, Interaction of temperament and environment in the production of behavioral disturbances in children. American Journal of Psychiatry, 1963, 12_0_, 142-148.
Clements, S. D. Task force one: minimal brain dysfunction in children. National Institute of Neurological Di's-orders and Blindness. Monograph No. 3, United States Department of Health, Education, and Welfare, 1966.
David, 0., Clark, J., & Voeller, K. Lead and hyperactivity. Lancet, 1972, 2_, 900-903.
Department of Health, Education, and Welfare. Report of conference of the use of stimulant drugs and the treatment of behaviorally disturbed young school children. Journal of Learning Disabilities, 1971, 4_(9) , 523-530.
Diagnostic and statistical manual of mental disorders (2nd ed.). Washington, D.C.: American Psychiatric Association, 1968.
Dixon, W. J., & Brown, M. G. Biomedical computer programs p-series. Los Angeles: University of California Press, 1977.
Douglas, V. I. Stop, look, and listen: The problem of sustained attention and impulse control in hyperactive and normal children. Canadian Journal of Behavioral Sciences, 1972, 4, 259-282.
II II
IJ
102
Dubey, D. Organic factors in hyperkinesis: A critical evaluation. American Journal of Orthopsychiatry, 1976, £6(2), 353-366. ^
Feiring, C. The influence of the child and secondary parent on maternal behavior. Dissertation Abstracts International, 1976 (Feb.), (8-B), 4225-4226.
Foshee, J. G. Studies in activity level. American Journal of Mental Deficiency, 1958, 62 / 882-886.
Friedland, S., & Shilkret, R. Alternative explanations of learning disabilities defensive hyperactivity. Exceptional Child, 1973, 4_0, 213-215.
Glennon, C. A., & Nason, D. E. Managing the behavior of the hyperkinetic child: What research says. Reading Teacher, 1974, 27 , 814-815.
Goggin, J. E. The role of temperament vis-a-vis milieu in the metamorphosis of hyperactive children. Paper presented at Southwestern Psychological Association (April), 1972.
Graham, P., Rutter, M., & George, S. Temperamental characteristics as predictors of behavior disorders in children. American Journal of Orthopsychiatry, 1973, 43_/ 328-339.
Henderson, A. T. , Dahlin, I., Partridge, C. R., & Engelsing, E. L. A hypothesis on the etiology of hyperactivity, with a pilot study report of related non-drug therapy. Pediatrics, 1973, 52_, 625.
Henderson, A. T. , Dahlin, I., Partridge, C. R., & Engelsing, E. L. Hyperactivity: The Henderson-Dahlin-Partridge-Engelsing hypothesis. Pediatrics, 1974, 54_, 515.
Hertzig, M. E. , Birch, H. G., Thomas, A., & Mendez, 0. A. Class and ethnic differences in the responsiveness of preschool children to cognitive demands. Monographs of the Society for Research in Child Development, 1968, 32/ 1-69.
Hetherington, E. Family interaction and psychopathology in children. In H. C. Quay & S. S. Werry (Eds.), Psychopatholoqical disorders of childhood. New York: John Wiley & Sons, Inc., 1972.
103
Huberty, C. J. Discriminant analysis. Review of Educational Research, 1975, 4_5(4) , 543-598.
Huessy, H. R. Study of the prevalence and therapy of the choreatiform syndrome of hyperkinesis in rural Vermont. Acta Paedopsychiatrv, 1967, 3£, 130-135.
Hunt, M. Slossen Intelligence Test. In 0. K. Buros (Ed.), The seventh mental measurements yearbook. Highland Park, New Jersey: Gryphon Press, 1972.
Kahn, D., & Gardner, G. Hyperactivity: Predominant diagnosis in child referrals. Frontiers of Psychiatry, 1975, 5_. 3.
Kerlinger, F. N. Foundation of behavioral research. Holt, Rinehart and Winston, Inc., 1973.
Kerlinger, F. N., & Pedhauser, E. J. Multiple regression in behavioral research. Holt, Rinehart and Winston, Inc., 1973.
Kirk, R. E. Experimental design: Procedures for the behavioral sciences. Belmont, California: Brooks/Coles Publishing Company, 1968,
Lachenbruch, P. A. An almost unbiased method of obtaining confidence intervals for the probability of misclassi-fication in discriminant analysis. Biometrics, 1967, 22, 639-645.
Lachenbruch, P. A., & Mickey, M. R. Estimation of error rates in discriminant analysis. Technometrics, 1968, 1_0, 1-11.
Lambert, N. M. , & Windmiller, M. An exploratory study of temperament traits in a population of children at rick. Journal of Special Education, 1977, 11.(1)/ 37-47.
Laufer, M. Longterm-management and some follow-up findings on the use of drugs with minimal cerebral syndromes. Journal of Learning Disabilities, 1971, 4_, 519-522.
Laufer, M., & Denhoff, E. Hyperkinetic behavior syndrome in children. Journal of Pediatrics, 1957, 5£' 463-474.
Laufer, M., Denhoff, E., & Solomons, G. Hyperkinetic impulse disorder in children's behavior problems. Psychosomatic Medicine, 1957, 19, 38-49.
104
Laybourne, P. C. Psychiatric response to the minimal brain dysfunction child. In R. P. Anderson Sc C. G, Halcomb (Eds.), Learning disability/minimal brain dysfunction: Research perspective and applications. Springfield: C. G. Thomas, 1976.
Lewis, D., Sachs, H., Bella, D., Lewis, M., & Heald, E. Introducing a child psychiatric service to juvenile justice setting. Child Psychiatry and Human Development, 1973, 1, 98-114.
Luria, A. R. The role of speech in the regulation of normal and abnormal behavior. New York: Liveright, 1961.
Marcus, J., Thomas, A., & Chess, S. Behavioral individuality in Kibbutz children. The Israel Annals of Psychiatry and Related Disciplines, 1969, 1_{1) , 43-54.
Marwit, S., & Stenner, A. Hyperkinesis: Delineation of two patterns. Exceptional Child, 1972, 38_, 401-406.
McDevitt, S. C , & Carey, W. B. The measurement of temperament in 3-7 year old children. Journal of Child Psychology and Psychiatry, 1978, 19_, 245-253.
Mclnerny, T., & Chamberline, R. W. Is it feasible to identify infants who are at risk for later behavioral problems? Clinical Pediatrics, 1978, 17_, 233-238.
Minde, K. , Weiss, G., & Mendelson, N. A five year follow-up study on 91 hyperactive school children. Journal of American Academy of Child Psychiatry, 1972, 1^, 595-610.
Mittleman, B. Motility in therapy. Psychoanalytic Study of the Child, 1957, 12./ 384.
Morrison, J. R., & Stewart, M. A. A family study of the hyperactive child syndrome. Biological Psychiatry, 1971, 2' 189-195.
Morrison, J. R., & Steward, M. A. The psychiatric status of the legal families of adopted hyperactive children. Archives of General Psychiatry, 1973, 22/ 888-891.
Myers, J. K., & Bean, L. L. A decade later: A follow-up of social class and mental illness. New York: John Wiley & Sons, Inc., 1968.
105
Ross, D. M., & Ross, S. A. Hyperactivity: theory, research, action. New York: John Wiley & Sons, Inc., 1976.
Rubin, R., & Balow, B. Learning and behaviors: A longitudinal study. Exceptional Child, 1971, 293-299.
Ryan, T. A. Multiple comparisons in psychological research. In R. E. Kirk (Ed.), Statistical issues: A reader for the behavioral sciences. Monterey, California: Brooks/ Cole Publishing Company, 1972.
Schmitt, B., Martin, H., Nellhaus, G., Cravens, J., Camp, B., & Jordan, K. The hyperactive child. Clinical Pediatrics, 1973, 12, 154-159.
Scott, W. C. Hyperactive children. British Medical Journal, 1973, 1, 113-114.
Senf, G. M. Model centers program for learning disabled children: Historical perspective. In R. P. Anderson & C. G. Halcomb (Eds.), Learning disability/minimal brain dysfunction syndrome. Springfield: Charles G. Thomas, Publishers, 1976.
Steward, M. Hyperactive children. Scientific American, 1970, 222, 94-98.
Steward, M. A., & Olds, S. Raising a hyperactive child. New York: Harper & Row, 197 3.
Tatsuoka, M. M. Multivariate analysis. New York: John Wiley & Sons, Inc., 1971.
Thomas, A., & Chess, S. Development in middle childhood. Seminars in Psychiatry, 1972, 4_(4) , 331-341.
Thomas, A., & Chess, S. Temperament and development. New York: Brunner/Mazel, 197 7.
Thomas, A., Chess, S., & Birch, H. G. Temperament and behavioral disorders in children. New York: New York University Press, 1968.
Thomas, A., Chess, S., & Birch, H. G. The origins of personality. Scientific American, 1970, 221(2)/ 102-109.
Thomas, A., Chess, S., Birch, H. G. , Hertzig, M. E., 5< Korn, S. Behavioral individuality in early childhood. New York: New York University Press, 1963.
106
Wender, P. H. The hyperactive child: A handbook for parents. New York: Crown Publishers, 1973.
Wender, P. H. Speculations concerning a possible biochemical basis of minimal brain dysfunction. International Journal of Mental Health, 1975, 4_, 11-28.
Werry, J. Developmental hyperactivity. Pediatric Clinic of North America, 1968, 15_, 581-599.
Williamson, G. A. The differentiation and treatment of two patterns of hyperkinesis in the classroom. Unpublished doctoral dissertation, Texas Tech University, 1978.
Williamson, G. A., Anderson, R. P., & Lundy, N. C, The ecological treatmient of hyperkinesis. Psychology In The Schools, 1980, 12, 249-256.
Zukow, P. G., Zukow, A. H., & Bentler, P. M, Rating scales for the identification and treatment of hyperkinesis. Journal of Consulting and Clinical Psychology, 1978, 46(2), 213-222.
All saints school
108
•;3o- 4-200 s c c e e c 7<i<i-nb-2 lUBBOCW, CeXAb -<5413
Dear Parents :Iay 2, 1979
The attached questionnaires are part of a research study Toeing conducted at All Saints School by Mr. Richard Langsdorf from the Texas Tech University Psychology Department. The ^^^eneral aim of this study is to better understand personality cnaracteristics in young school-age children, both normal and emotionally distressed, rto-. Langsdorf is asking the parents of children at our school to participate in this st-jdy as his group of normal children.
If you are^interested in participating we would like you to be involved in one of two ways. You may simply fill out the questionnaires and have your child ret^am them to his or her classroom teacher. It should take you about 30 minutes to complete these forms.
The second way you may want to participate in this study is to complete the enclosed questionnaires and have your child return them to the teacher and in addition, give your permission for your child to be given three tests. These tests would include a brief measure of intelligence, a test of visual-motor coordination aind a Vigilance Test. The Vigilance Task measures your child's ability to pay attention. The child is asked to attend to a changing pattern of flashing lights, pressing a button whenever a specific light pattern appears. During the Vigilance Task, level of physical activity is also measured. The entire testing will take a little more than one hour and will be conducted during regilar school hours o.n the campus of All Saints.
As part of the study you will be given the results of the tests either by telephone, or by scheduled appointment with Mr. Langsdorf and a member of the school if you elect to have your child tested. We feel the results of this brief evaluation could provide both you and the school with information to better understand your child.
If you decide to participate, please read and sign the enclosed consent form and return it together with the completed questionnaires. Each resesirch study at TTTJ is required to have the informed consent of the participant and in particular, the portion referring to insurance is required for every research project. You should be aware that there sliould be_ no risks or discomfort for you or your child should you decide to participate.
If you decide not to participate, could you please send the materials back to your child's teacher. Thank you for your interest.
Richard Lang^dori ,}[.S, \ Researcher
Betty-^. Sny-±^ Director ' / /C
-JC/X^^— / A» A- Robert ?. A.iderson, Ph.D.
112
Hyperactive Group (N=20)
Code Name
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
Chair a
Count
445
532
265
729
376
372
528
272
305
632
377
796
897
727
348
308
904
585
445
168
Nonhyperactive Clinical Group (N=20)
TOC Code Name
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
Chair Count
16
227
15
187
93
0
47
263
110
13
90
43
161
312
412
257
105
167
31
275
TOC
0
5
0
0
120
0
0
60
285
0
5
0
0
0
0
90
0
0
0
0
70
200
610
356
725
445
662
1468
395
0
645
21
0
675
625
1130
90
180
0
1184
113
Code Name
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
Chair Count
266
294
38
358
527
234
99
144
270
67
104
676
874
8
336
265
TOC^
405
0
0
0
265
0
0
0
0.
0
0
0
0
0
0
461
Control Group (N=31)
Code Name
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
Chair Count
334
53
156
0
775
34
639
23
334
207
25
142
0
68
270
TOC
81
692
0
0
10
0
0
0
0
0
0
180
0
0
0
stabilimetric activity count in raw score units
Time-out-of-chair count in seconds
115
ATTENTIONi Q u e s t i o n n a i r e s must be comple t ed by mother of c h i l d
INFORiVATION SHZ2T
C h i l d ' s Name Sex
Date of C h i l d ' s B i r t h - ^ ^ ^ a S y - y i a r ^'^'""^ " ^ e .
School Grade
Rater's Name
Occupation of Father Kothe;
Telephone number
Is your child currently being treated problems? j | j I (check one)
for emotional or behavioral 1 (check one)
yes no
I do do not give my permission for my chi ld to pa r t i c ipa t e in the t e s t i n g par t of t h i s research study.
Signature of parent or lega l guardian
BEHAVIORAL STYLE QUZSTIONTTAIRE
I n s t r u c t i o n s :
1. P lease base your r s c i n g on the c h i l d ' s r e c e n t and z-rr^r.z behavior (:'.-.e l ^ s : four to s ix weeks) .
2. Ccr.sider only vpur ovn Impress ions and obser ' /a t ic r . s of tp.e c . - i id .
3. Rate each q u e s t i o n Inde re rodent Iv . Do not purposely a t t * - ; t to preser.t a c o n s i s t e n t p i c t u r e of the c h i l d ,
'•*• Use e x t r e - e r a c J - g s where a p p r o p r i a t e . Avoid r a t i n g only r.sar the r . iddle =f Che s c a l e .
5. Rate each Iteai cuicklv.- I f you cancoc d e c i d e , s k i s the itesa and co;r.e back to ic l a t e r .
S- Rate every iter:;. C i r c l e the nur.b«r cf any I t e n t ha t you are unable t= ^rsver due to lack of i n f o r r . a t i o n or any i ters t h a t does not apply to your tr.-.ld.
116
USING 711?: SCALE SHCWN QELOU'. PLEASE V RK AN "X" 'N T-T. SPACE ; FTEN THE CHILD'S RECENT AND CUR.RENT 5EHAVICR -J\S 3EEN L : : < Z T - ! : II BY EACn ITEM.
Almost Rarely Usually Usually
never does not docs
1 2 3 i
1. The child is moody for nore than a few
minutes when corrected or disciplined.
2. The child seems not to hear when involved
in a favorite activity.
3. The child can be coaxed out of a forbidden
activity.
U. The child runs ahead when walking with the parent.
5. The child laughs or smiles while playing.
6. The child moves slowly when working on a
Frequent 1y
5
almost
never
aImos t never
almost
p.ever
a-.TICS t
never
almost never
almost
1
1
1
— : — •
1
*
2
2
"1"
2
~T
A l m o s t
a l w a y s
5
3 U
3
~ T"
1 T"
~3 r
5
5.
C
^
~r
6
0
—:—
0
3Incs:
alway;
al.Tcs t always
al.-Tics t always
al-cs:
2 -Wa'-":
a.T.cs: always
al-c3:
project or activity, never 1 2 3
7. The child responds intensely to disapproval, almost : : : :___:
never
0 3.ways
i.T.cs: 6 a 1 •-• a •• 3
5. T'-.e child needs a period of adjustment to
get used to changes in school or at home.
9. The child enjcys games that involve
running or junjping.
10. The child Is slow to adjust to changes in
household rules.
11. The child has bowel movements aC about the
same time each day.
12. The child is willing to try new things.
13. The child sits calmly while watching TV or
listening to music.
1^. The child leaves or wants to leave the
table during meals.
15. Changes in plans bother the child.
almost : :
never 1 2 3
almost : : never 1 2 3 -k 0
al.most
never 1 2 3
almost : never 1 2
almost :
never 1 2
almost :
never 1 2
almost : never 1 2
almost ;
never 1 2
i* 5
3 U 5
:i U !) i
3 _;
3L 1"
3 ^
a 1
3 1
al
z \
a 1
ai al
a a
a a
3
"C 5
-•ay"
-OS:
-• a >• 3
-cs :
w a y 3
-cs: ways
mos :
w a >• 3
.mcs t
.ways
..TCS t
Iways
1-cs:
16, The child notices minor changes in mother's almost : dress or appearance (clothing, hairstyle, e t c . ) . never 1
i.ways
aires : a.wi-. s
P - • (T7
117
Almost never
1
'.arely Usually does not
3
' sual ly does 4
•equen .-••, ; ~ c < ; :
always 6
17. The child does not acknowledge a call to
come in if involved in something.
18. The child responds to mild disapproval by the parent ( a frown or shake of the head).
19. The child settles arguments with playmates
within a few minutes.
:0. The child shows strong reaction to things,
both positive and negative.
21. The child .ad trouble leaving the mother
the first three days when he/she entered
school.
22. The child picks up the nuances or subtle
ties of parental explanations (example: implied
meanings) .
23. The child falls asleep as soon as he/she is
put to bed.
24. The child moves about actively when he/she
explores new places.
25. T h e c h i l d likes tc go to new places r a t h e r
Chan f a m i l i a r o n e s .
26. The child sits quietly while waiting.
27. The child spends over an hour reading a
book or looking at the pictures.
23. The child learns new things at his/her
level quickly and easily.
29. The child smiles or laughs when he/she
meets new visitors at home.
30. The child is easily excited by praise.
31, The child is outgoing with strangers
32. The child fidgets when hs/sze has to
stay still,
33. The child says that he/she is "bored" with his her toys and games.
aImos t
never
al.most never
2 3 ^ 5 aiwav
I 2 3
almost
never 1 >
almost never 1
alr.cs:
,-,ever
a-most never 1
almost
never 1
almost never 1
almost
never 1
alm.ost never 1
al.mcst
never 1
almost never 1
aLmost
never 1
5
6
6
-
c
;
6
6
5
6
•3
•:
al-cs: always
a.mo s t
always
a1-cs: alwa:--s
alr-cs : 31 w a y 3
il.TCS :
a 1 ••• a ;.• s
a l m c s :
a 1 wa'-'S
-1 .T c s : alway s
al-.os: always
aires : always
aires :
always
a1-cs: a-wavs
a l.mo s t
always
almost : : : : : almost never 1 2 3 4 5 6 always
almost
never 1
almost
never 1
almost
never 1
' a-res:
6 always
a Ir.c 3 :
-. alwavs
alr.cs:
•i always
• f
A I.mo.St
never 1
Rarely
2
Usually
does not
3
U s u a 11 y
does 4
"requencly
3
A l.Tos:
alwavs /-«
118
3 i . The c h i l d i s annoyed a: Ln te r r ' . i p t ing p l ay almost : : : : : almost :o coTTipiy w i th a p a r e n t a l r e q u e s t . r e v e r 1 2 3 >- 5 6 always
35. The c h i l d p r a c t i c e s an a c t i v i t y u n t i l h e ' s h e almost : : : : : a lmost m a s t e r s i t , never 1 2 3 4 5 6 always
36. The c h i l d e a t s about t h e same amount a t a lmost : : : : "• a lmost s u p p e r from day to day . never 1 2 3 4 5 6 always
37. Unusual n o i s e s ( s i r e n s , t h u n d e r , e t c . ) a lmost : : : : • a imcst i n t e r r u p t the c h i l d ' s b e h a v i o r . never 1 2 3 4 5 6 always
3S . The c h i l d compla ins when t i r e d . al.r.ost : : ; : : a . r . c s : never 1 2 3 4 5 o aiways
39. The c h i l d l o s e s i n t e r e s t in a new toy or a lmost : ; : : : a - r c s : ga.me the same d a y . never 1 2 3 ^ 5 6 always
40 . The c h i l d becomes e n g r o s s e d in an i n t e r - a lmost : : : : : a lmost e a s t i n g a c t i v i t y for one h a l f hour or more . never 1 2 3 4 5 6 a ivays
4 1 . The c h i l d c r i e s i n t e n s e l y when h u r t . a lmost : : : : : almost never 1 2 3 4 5 6 always
42 . The c h i l d r e a c t s s t r o n g l y to k i d d i n g or a lmost : : : : •___ almost l i g h t - h e a r t e d c e m e n t s , never 1 2 3 ^ 3 6 alwavs
43 . The c h i l d a p p r c a c h e s c h i l d r e n h i s / h e r age a lmost ; : : : : almost t h a t h e / s h e d o e s n ' t know. never I 2 3 ^ 5 ^ alwavs
44. The c h i l d p l a v s q u i e t l y w i th h i s / h e r toys a lmost : : : : : almosc and games. "«"«^ 1 2 3 4 5 6 always
4 5 . Tne c h i l d i s o u t w a r d l y e x p r e s s i v e or h i s / h e r a lmost : : : : : almost »m«-<«n< ^ever 1 2 3 4 ^ 6 always e m o t i o n s . 46. The c h i l d i s e n t h u s i a s t i c when h e / s h e a lmost : : : • • a lmost mas t e r s an a c t i v i t y and wants to show never 1 2 3 ^ 5 5 always
e v e r y o n e .
47 . The c h i l d i s s l e e p y a t h i s / h e r b e d - t i m e . a lmost : : • • • ^^ a lmost never 1 2 3 4 ^ 6 always
48 . The c h i l d s t o p s an a c t i v i t y b e c a u s e some- a lmost : : : :___: almost t h i n g e l s e c a t c h e s h i s / h e r a t t e n t i o n . never 1 2 3 >* ^ 6 a.ways
49 . The c h i l d i s hungry a t d i n n e r t i n e . a lmost : : : ••..—. a-~<=s. never 1 2 3 4 : ^ always
50. The c h i l d h o l d s back u n t i l s u r e of h i m s e l f / alm.ost : : : : : a . r . c s t h e r s e l f . never 1 2 3 4 5 6 a.wavs
119
AImos t never
1
Rarely
2
Usually does not
3
Usually does 4
Frequently A:mcs: alwavs
51. The child looks up when someone walks past almost : : : : ; almost tlie door-way. never 1 2 3 4 5 i nlwayj
52. The child becomes upset If he/she misses a almost : : : : : almost regular television program. never 1 2 3 4 5 6 always
53. The child reacts strongly (cries or com- almost : : : : : almost plains) to a disappointment or failure. never 1 2 3 4 5 6 always
54. The child accepts new foods within one or almost : : : ; : ai.most two tries. never 1 2 3 4 5 6 always
33. The child has difficulty getting used to almost : : : : : almost new situations. never 1 2 3 4 5 6 always
56. The child will avoid misbehavior if almost ; : : : : almost punished firm.ly once or twice. never 1 2 3 >* 5 6 always
57. The child is sensitive to noises (tele- almost : : : : : alm.ost phone, doorbell) and looks up right away. never 1 2 3 4 5 6 always
53. The child prefers active outdoor play to almost ; : : : : alm.ost quiet play inside. never 1 2 3 4 5 6 always
59. Tne child dislikes milk or other drinks almost : : : : : alm.ost if not Ice-cold, never 1 2 3 4 5 6 always
60. The child notices differences cr changes almost : : : : ; almost in the consistency of food, never 1 2 3 4 5 6 always
61. The child adjusts easily to changes in almost : : : ; : almost his/her routine. never 1 2 3 4 5 6 always
62. The child eats about the same amount at almost : : '. : • a-mcst breakfast from day to day. never 1 2 3 4 5 6 always
63. The child seems to take setbacks in almost : ; : : • almost gcj-^de, "«''«' 1 2 3 4 5 6 always
64. The child cries or whines when frustrated. almost : : : : : almost never 1 2 3 4 5 6 always
65. The child repeats behavior for which he/she almost : : : ; :__ almost has previously been punished. never 1 2 3 4 5 ^ always
66. Z^e child looks up from playing when the almost : : : : '• almost , u 4 never 1 2 3 4 5 6 always
telephone rings. never i ^ J 67. The child is willing to try new foods, almost : : : : : almost
never 1 2 3 4 5 6 always
120
Almost never
1
Rarely
2
Usually
does not
3
Usually does >«
.- requen: .y r .mcs : always
3 5
68. The child needs encouragement before he/she almost : : : : : almost will try new things. never ~I 2 3~ - 5 •: alwavs
69. The child cries or whines when ill with a almost : : : : almost cold or upset stomach. never 1 2 3 -» 5 6 alwavs
70. The child runs to get where he/she wants to almost ; : ; ; : almost go. never 1 2 3 4 5 6 always
7 1. Tlie child's attention drifts away or lapses almost : : ; : : almcs: when listening to parental instructions. never 1 2 3 4 5 0 always
72. The child becomes angry with one of his'her almost : : : ; : almrst playmates. never 1 2 3 - 5 6 alwavs
73. The child is reluctant to give up when almost : : : : : aires: trying to do a difficult task, never 1 2 3 - 5 6 a-wa-s
74. The child reacts to mild approval from the almost : : : : : al-cs; parent (a nod or smile), never 1 2 3 4 5 6 alw2-/s
75. The child requests "something to eat" be- almost : : : : : al.most tween meals and regular snacks. never 1 2 3 - 5 6 alw ..-
76. The child rushes to greet the parent or almost : : :j : : almc?: greets loudly after absence during the day. never 1 2 3 4 5 6 always
77. The child looks up when he/she hears voices almost : : : : : al.mcs: In the next room. never I 2 3 -» 5 6 always
78. The child protests when denied a request by almost : : : : : almost by the parent. never 1 2 3 4 5 6 always
79. The child ignores loud noises when reading almost : : : : : a-ros: or looking at pictures in a book. never 1 2 3 4 5 6 always
80. The child dislikes a food that he/she had almost : : : :__:__ almost previously seemed to accept. never 1 2 3 4 5 6 always
81. The child stops what he/she is doing and almost : : : : :__ almcs: looks up when the parent enters the room. never 1 2 3 4 5 6 aivays
82. The child cries for more than a few minutes almost : : : : : a.most when hurt. never 1 2 3 4 5 6 always
83. The child watches a long ( 1 hour or more) almost : : _ _ : _ _ : _ _ _ : ^ almost
TV program without getting up to do something never 1 2 3 4 5 6 always
else,
84. The child spontaneously wakes up aC the almost : : : : :___ a..-.cs-
usual time on weekends and holidays, never I 2 3 - 5 e a.wa-s
121
Almost never
Rarely Usually does not
3
Usually does 4
Freque: A.mo s :
aIwavs
35. The child responds zo sounds or noises unrelated :o his/her activity.
36. Tbc child avoids new guests or visitors
a.most : : : : never 1 2 3 '* 5
almost : :
never 1 2 3
a_mes: n 1 w 3 •' s
: almost
5 6 nIwavs
57. Tlie child fidgets when a story is being read to him/her.
almost : : : : : almcs: never 1 2 3 4 5 6 always
5S. The child becomes upset or cries over minor falls or bumps.
39. The child interrupts an activity to listen :o conversation around him/her.
almost : : : never 1 2 3 4
almost 6 alwa-s
a L.-cs t never I
a -mos: aIwavs
90. The child is unwilling to leave a play activity that he/she has not completed,
91. The child is able to fall asleep when there is conversation in a nearby room,
92. TV.e child becomes highly excited when pre
sented with a new toy or game.
93. The child pays attention from start to
finish when the parent tries to explain some
thing :o him/her.
94. rne child speaks so quickly :hat it is some
times difficult to understand hi.m/her.
95. The child wants to leave the table during
meals to answer the doorbell or phone,
96. The child complains of events in school or
with playmates that day.
97. The child frowns when asked to do a chore
by the parent.
98. The child tends to hold back in new
situations,
99. The child laughs hard while watching
television cartoons or comedy.
100. rne child has "off" days when he/she is
moodv or cranky.
almost never 1
aimos t
never
almost never
almost never
almost
never I
almost
never
almost
never I
a--cs: 6 always
al-ost 6 always
al-ics: 6 always
.1 I.mos: 5 alwavs
3
: aires:
5 5 always
: almost 5 6 alwa',-'3
: almost 5 6 alwavs
almost : : : : : almost never 1 2 3 4 5 6 aivays
: ; : : almost
T* 3 4 5 6 always
: : : aires:
T~ 3 4 5 6 always
: : : : al-cst ~2~ 3 4 5 6 a-ways
aImos t never
almos t never
almost never
1
1
i
BEHAVIORAL STYLE GUESTIOMNA) RE S C . „ H C : H , „
m^-
122
H A M ( DATE or • A [ IMG t.Oi
S- ' =i a ;; 2 .- -. ; • - » -" r;; « o -
(nstructions to scorer: 1) Check off above responses on questionnaire. 2) For category score add checks in 6 columns I multiply SLIDIS bv factor. Resulting products are added S divided by number of items rated, productmg category score.
< - 1 i ?
" I I . . I I - 1 J
1 * t 1 '^
-• • 1 — ,
.
1
f '
1 »
, • • 4
; - 1 '.
." ! • — »
j *
- j :
zii
' 1 ! 1 ' 1 <
1 . a
1 ' •* 1 1 <^
1 1 > i i '
• t »
' - ! ! I- '
1 , i ^ t *
• 1 c
> • 1
* 1 *
1 - . • -( i 1 - 1 • -
« 1 . - -
u < ' w -
» 1 ^ •• .« 1 , ^ m
^ ' y. »
' - ' -1 - 1 '... - M < u 1 u , - X
' • 1 . . k
, •• •
, -i -
,-i -
•
: - - -• > • « 1 ! ! »
I - • 1 ^ «« 1 * *
1 . -1 - -
—I : - 1 . S
. -
1 - 1 1 -1 1 »
- - 1 1 -* 1 1 •
- 1 ; s -A i
-fen - . - J
1 1 .
w
t M
3
-^ -» -»
3>
9
-• • -
* ^ U
» J i
n
J 3
--u
•. < » .:; --' * •.» ^
:3 A
-^ M
--
J a a
». ^ ^ «4 U W
* * k
^ >« t«
• • « 3
M
W
-
3 ---• * > a
.:: -.. u
» v„ 5 1
j ;
--' ^ \* o.
^ ^ -
M •«
W W
--
3 A
J :? * 4
- ^ » >-• u * »• w>
- • »
:; -M
W
* ^ a »
• — «• •* » w u *
— •
0 » -A
-•*
5 ^
*<• t «
«-• ^ » u»
<A •«
» — » .a .3
- a >
w ^ ^
u . - *
* w w
.J* •>• w
» - -
->« ** b> w
^ Jk
« «« a «
i; --u
« U i
• ^ -..• • •
« ---» •• J»
J
a
--• ---
i
,>. ^ -"
^ • »
-» •-•
V .
-» 3 3 —
Ui . . .
• U
b . ^
- > • 3
J ,
--U
» -a
;i --
• - >
--3 .
t -— -* -•
;; -~ w
•
. •« .. * -M
-• ; -». * J. .*
- -u ^
w „
- -w
'.> ^ u
« -
- - . _ 1-, w - . . .
A A * *
. ^ - . J» . .
J l J. 0, >
3 a
i «
• ---
a -.. w
* l «
91
J
--W
» --•
-3*
^ •. U
.. -
9
>.* . ' ~ -
--S .
^ ^ .. s
w . .
u u
. > ^ -9 ^
V. U. ~
h . » ^ .
, . W A
JI . . ^
» - •
-_ » (.. A
-
A »
J) —
</. - » *., •_ » . . o .
- :. :;; 7»
^ » M
--
3» —
u . - J
• ' J
(., * u <„
- -• 3
n
J.
> L ,
--
T
--C
* a
3.
^ 9
^ • U
U
-^ 3 »
.« » <.. w
-
-u
. • "
; J — 7
.. .. - -. J
0, ^
^ -* : ^ ----11
.• • •
-w
-J •* 5
» .- .-. "^ • .
- c . - ,
- » » - . w ^
- -71 ^
• • •
» W
-. -
i S
_ _ * * " -
o -a » '
- ^ - . -- - - v 1 t
,, - _ . , ; •> - - • , , - 1 J . .J ^ ' .
- - -:
,-; -
' ' '•i : 1 ;l
; ' ' • -
•• ' 1 3 - : -1
^ -" ^ !! - V ; •_!
» ^ . 1 -1 ^ ~. • '• - \
. : ) * * 1 ; - '
^ ~ - 1
, !" "" ; i " 1 J . " •
^ - 1
9 — 1 • ' . 1
^ - . • ci . _ . '. - 1
-• *' : .. ' ' : - , . . " ' 1
: Z \ - . 1
1 ^ '
i 1
•J 1 o
9* 1 ' ^
" M ' - • •
: • • 'r '_
1 1° . 1
3g..-L\VI0RAL 3TYL?: CUESTIJ' ." . 'AI?." •=rori.
f o r 3 t o 7 7B«r o l d c h l l d r e a
Oeve loped (1975) by Saam C. MoDsTltt , ? h . D. 4 W i l l i / , . 3 . ':»rsy,>«.D,
C h i l d ' s N'AM : i , t» of ?.«»tiii«
B o n t h s . Sax Age a t r n t l B ^ i . y e a r s ,
Catecory s c o r a f r o a Scorln;? SheaCi
123
P r o f i l e I Place aark In apDro^riate bor bslcwj
n.o S . D .
MEAN
• 1 . 0 3 . D
.^ctlTirs-
h i g n
3 . 5 6
2 . 8 1
low
.^hytha.
arr .^ t .h
3.^*3
2 . 7 5
2 . 0 7
v e r y r h y t h .
A?p/Vtt.h
v i t h d r -
3 . 9 3
2 . 9 9
2 . 0 5
app .
A i a p t .
s l o w l y
3 . 2 7
2 . 5 5
1 . 8 3
Tsry a d a p t .
I n t a « 3 .
Intaasf l
5 . 1 7
' i .52
3 . 3 7
• i l l
"ocd
n^sra : ! 7-?
3 .99
3 .31
2 . 6 ;
p o s l t i T e
P-irsl^t.
r. on per 3
3 . 5 6
2 . 8 7
1 1 Q
hl?;h p' r
!?l5 tr". c 1
i l 3 : r ° . c
4 . " c
3 .35
3 ,08
". 0 n -i l s t r a c
Ohr-.in.
low
4 . ;3
3.:^3
3. :;3
h l s n
Diagnostic Clusters
easy
Dilf
•Slow low
rhtha.
arrythm.
app. N >'«*«Mn^ *&«•
wit.hdr.
wlthdr.
adapt,
3lo"wl7 adapt. s lowiy .adaat.
a l l d
l a t ensa
a l l d
p o s i t i v e
nec-atlv!
aagatl^e
D«f l» l t l on of d i a g n o s t i c c l u s t e r s used for ladUti-duai 3corl;i2;t f a s y ^ ^ J c S r a - ^ a a c a r !nan aaan In no . o r e than two of ^^'^''t^^^nai^hrr z - e ' 'ar
( r h y t h a l c l t y . approach, a d a p t a b i l i t y , Inte.nslty 4 aood) and nelthrsr z . e a . a r than one standard d e v i a t i o n . . ^-. , . . / ..„ „.*..^«-^i.- <«•« Abovai
a f f l c u l t - U or 5 scores grea ter than -ean in d i f f i c u l t / e a s y categor ies tAsaoova; " h i s -U3t include I n t e n s i t y and two scores greater than one standard devla Ion
S l o w ' ^ i w J J i u i " as def ined ab iva , but I f e i t h e r withdrawal or slow ;<i-Pt;^i l i^7 I s V a a t e r than one standard d e v i a t i o n , a c t i v i t y nay vary up to 3.:.3 and aood
> t e r n ^ d i ; S - ' : n ^1ha' ;^! - ln ter» .d la te h i g h - ^ or 5 ^ / ^ / " ^ ^ ^ / ^ f ^ r : ^ T " ^ mean with one >1 standard d e v i a t i o n , or 2 or 3 *bove nean v l - h . or 3
>1 standard d e v i a t i o n . Intermediate low- a l l other I n t a n a e d i a . e s .
T h l 3 c h i l d ' s d iagnos t i c c l u s t e r Coaaentsi
Date of scorl.ng
Scored try
125
Parent Rating
SOMETIMES
Behavior Form
Instructions: Circle the answer which best fits your child's da^
to-day behavior.
1. Unusually hyperactive: HOME SCHOOL
2. Jumps from one activity to another: 3. Short attention span:
4. Fidgets:
5. Is unpredictable, unmanageable:
6. Irritable:
7. Overly sensitive:
8. Quick tempered, explosive:
9. Panics easily:
10. Tolerance for failure and frustration is low:
11. Emotionally high strung:
12. Told ahead of time about an outing or appointment becomes anxious or upset:
13. Exceptionally clumsy:
14. Poor coordination:
15. Eyes and hands don't seem to function together
16. Has trouble buttoning:
17. Has trouble drawing, writing:
18. Was slow learning to walk:
19. Trouble with bicycle:
20. Trouble catching ball:
BOTH
Yes
Yes
CONSTANT
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
NO
No
No
No
No
No
No
No
No
No
No
No
No
No
No
No
No
No
No
N'C
21. Speech development has been slow: les No
-^a^ ~ - ^ - • • Li<.
^
_^-i
22. Speech is not clear:
23. Reacts adversely to changes in routine:
24. Can't seem to keep from touching everything and everyone around him:
25. Not learning in school although seems "bright"
26. Is child lazy—not "trying" to do well in school:
27. Daydreams while doing homework assignments:
28. Knows work orally at home—gets to school and has to write it down—fails miserably:
Yes
Yes
Yes
Yes
Yes
Yes
126
No
No
No
No
No
No
Yes No
I *Z 3
128
I would like you to participate in a study being con
ducted in the Psychology Department at Texas Tech University.
The primary aim of this research project is to examine the
relationship between certain temperament characteristics in
young school-aged children and the presence or absence of
emotional or behavioral problems.
If you decide to participate in this study with your
child, you will be asked to complete a form requesting back
ground information on your child and family, and two ques
tionnaires about your child. One questionnaire asks you to
characterize your child's behavior in different kinds of
everyday situations. On the second questionnaire you will
be asked to rate your child on a list of 28 problem areas.
While you are completing these questionnaires your son or
daughter will be given three different tests including a
brief measure of intelligence, a test of visual-motor coor
dination and a Vigilance Test. The Vigilance Task measures
your child's ability to pay attention. The child is asked
to attend to a changing pattern of flashing lights, pressing
a button whenever a specific light pattern appears. This
test lasts 35 minutes. During the Vigilance Task, level of
physical activity is also measured. The entire study should
take about one and one-half hours.
129
After you have participated in this study I will tell you
exactly what the reasons are for doing the study, what the
expected relationships between the child's temperament and the
presence or absence of emotional or behavioral problems might
be, and what exactly I hope to learn from this study. At this
time I will give you a summary of the results of the evalua
tion and we may discuss any concerns you may have pertaining
to these results.
I will retain the results of the evaluation in a confi
dential file accessible only to myself and my faculty advisors
for this study. If you wish to have a summary of the findings
sent to your child's pediatrician, school or other concerned
party, it will be necessary for you to sign a separate release
of information form.
It is important that you realize that these questionnaires
and tests are the same ones which were part of the routine psy
chological evaluation already completed on your child. I am
simply asking your permission to use the information gathered
from this routine evaluation as part of a research project.
You should also be aware that there should be no risks or dis
comfort attendant upon your participation in this study.
If this research project causes any physical injury to
you, treatment is not necessarily available at Texas Tech
University or the Student Health Center, or any program of
insurance applicable to the institution and its personnel.
130
Financial compensation for any such injury must be provided
through your own insurance program. Further information
about these matters may be obtained from Dr. J. Knox Jones,
Jr., Vice-President for Research and Graduate Studies, tele
phone 742-2152, Room 118, Administration Building, Texas
Tech University, Lubbock, Texas 79409.
I understand that I may not derive therapeutic treatment
from participation in this study. I understand that I may
discontinue this study at any time I choose.
Signature of Subject: Date:
Signature of Investigator: Date