Copyright by RICHARD C. LANGSDORF 1980

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.

APPENDIX A: LETTER TO PARENTS

107

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.

APPENDIX B: PICTURE OF VIGILANCE

TESTING APPARATUS

109

• • I

110

B«!

APPENDIX C: RAW SCORE DATA FOR TWO MEASURES OF

ACTIVITY LEVEL FOR THREE COMPARISON GROUPS

111

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

APPENDIX.D: BEHAVIORAL STYLE QUESTIONNAIRE

114

a!«

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

APPENDIX E: ZUKOW HYPERKINESIS RATING FORM

124

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

APPENDIX F: PARENT CONSENT FORM

127

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