rt.-lllllll|

 

lllgllllllflllzllflfllflllfllljflfllllflW

Mid‘zigan Stat:
University

 

This is to certify that the

thesis entitled

A STUDY OF THE RELATIONSHIP BETWEEN TNFANT
TETTPFRAMFNT VARIABLES IN BIOLOGICAL
RISK AND AVERAGE POPULATIONS AND
THE SELF-CONCEPT OF THE

“0 7 ,
preserritfefi by

Gary B. Saltzman

has been accepted towards fulfillment
of the requirements for

Ph.D. degree in Educational l.Jsvcflaoloszv

 

 

Q—«QW A

Major professor

Date__é. ' C1" 7 8

0-7 639

 

<:) Copyright by
GARY BRYAN SALTZMAN

1978

A STUDY OF THE RELATIONSHIP BETWEEN INFANT
TEMPERAMENT VARIABLES IN BIOLOGICAL
RISK AND AVERAGE POPULATIONS AND
THE SELF-CONCEPT OF THE
MOTHER

By

Gary Bryan Saltzman

A DISSERTATION

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

DOCTOR OF PHILOSOPHY

Department of Counseling, Personnel Services
and Educational Psychology

1978

ABSTRACT
A STUDY OF THE RELATIONSHIP BETWEEN INFANT
TEMPERAMENT VARIABLES IN BIOLOGICAL
RISK AND AVERAGE POPULATIONS AND

THE SELF-CONCEPT OF THE
MOTHER

By
Gary B. Saltzman

Increasing attention has recently been devoted to the
study of infants, their competencies, differences, and
interactions with their caretakers. The present study was
concerned, in particular, with infant temperament, and the
variations in infants' styles of behavior in relation to
several factors that were believed to merit specific consi-
deration.

Today, many infants are surviving pre-, peri-, or post-
natal trauma who would not have done so in earlier years.

As a result, more babies are ending up in "high risk" popu-
lations, which means that they are more vulnerable for
developing problems capable of interfering with normal growth
and development. A question explored at this time was
whether infants in this "biological risk" group develop more
difficult temperament styles than those infants born without
complicating circumstances.

Also examined was the reciprocal and mutually reinforcing

interaction between mothers and their infants. Specifically,

Gary B. Saltzman
relationships between infant temperament and the self-concept
of the mother were explored, as were the mother's observa-
tions and perceptions of her child's temperament. Finally,
temperament was studied in relationship to 1) the sex of the
child, 2) the child's ordinal position in the family, and
3) the developmental level of the infants in the two popu-
lations in this project.

The sample in this study consisted of 150 mother/infant
dyads. Seventy-five of theses pairs came from a "high risk"
population of infants seen at a developmental assessment
clinic. The remaining seventy-five pairs were contacted
through "well baby" clinics, and for the purposes of this
study were considered the "average" population.

Mothers were requested to complete both the Carey/
McDevitt infant TemperamentQuestionnaire, which was used as
a relatively objective and detailed description of their
infant's temperament, and the Tennessee Se;f:Concept Scale.
In addition. mothers and psychologists each completed a form
on which they recorded their general impressions of the
infant's temperament. Infants in both populations were
administered the Qenver Developmental Screening Test.

In analyzing the data in this study, the chi square
statistic was used to examine the relationship between tem—
perament, biological risk status of the child, and the above
stated variables. The effect of both the temperament of the
infant and the infant's biological risk status on the self-

concept of the mother was examined using a twO-way analysis

Gary B. Saltzman
of variance.

Several major findings emerged. One, there was no sig-
nificant relationship found between temperament and the risk
status, the sex, and the ordinal position of the child in the
family. Two, statistical significance was found in the rela-
tionship between scores on the Denver Developmental Screening
Test and the child's temperament. Infants with developmental
problems were found more often in the difficult temperament
clusters. Three, the self-concept of mothers of infants in
the difficult temperament categories was noted to be signifi-
cantly lower than that of mothers of average babies. Four,
mothers of infants at "biological risk" also had lower self-
concepts than mothers of "average" babies. Five, discrep—
ancies were found between mothers' perceptions, psychologists'
perceptions, and ratings on the more objective and detailed
Infant Temperament Questionnaire. And six, it was observed
that mothers of "difficult" temperament infants were more
likely to perceive their child as "easier" than both the
psychologist and the questionnaire would indicate.

Rather than cause and effect, this study has been res-
tricted to demonstrating associations between the stated
variables. Nonetheless, these associations may be of impor-
tance for the prevention of future behavioral disorders in
the child and emotional problems in the mother. In.parti-
cular, the lower self-concept of mothers of difficult and

high risk infants was felt to be especially significant.

Gary B. Saltzman
This, along with the discrepancy between mother's perceptions
and the psychologist's and questionnaire evaluations of the
infant's temperament, was believed to be suggestive of poten-
tially stressful mother-infant relationships. Keeping track
of those mother-infant dyads where there is a difficult infant
and providing supportive services at various intervals when
disharmony is apparently evident were suggested as possible
measures for preventing more serious problems from developing.

The importance of recognizing individual temperament

differences in infants and the nature of the infant's rela-
tionships with his parents was emphasized, as was the impor—

tance of parents sharing the responsibility of child rearing.

Dedicated to Mindy....Who helped me to realize that

"When you dream, dream big!"

iii

ACKNOWLEDGMENTS

It has taken me a considerable amount of time to write
this brief section simply because it is so difficult to put
down in words the gratitude and appreciation I feel towards
all those individuals who have actively helped me in complet-
ing this project.

Attempting this study has been quite an adventure for
me. Not only did I learn about the topics that I explored,
but I also learned much about myself. With each further step
in the journey towards completing this work, I feel as though
I have grown a little. I guess this expresses my overall
sentiment towards the time I have spent as a doctoral student
here at Michigan State. Many people have contributed to this
growth, and in fact have helped me to realize educational and
professional objectives that only a few short years ago were
simply dreams.

Much of the success of this dissertation and fer that
matter, my own personal and professional growth, has come out
of the knowledge, guidance, and support of various people who
I have been indeed fortunate to work with at various stages
in this project and in my educational career at M.S.U.

Foremost among these individuals is the Chairman of my

Doctoral Committee, Dr. Don Hamachek. From the very beginning

iv

of my studies at Michigan State, when he served as my advisor,
right through to the completion of my dissertation, he has
consistently been there when needed and has always been will-
ing to share his knowledge and experience. The interest he
has shown in my educational and professional endeavors and
the support he has provided have made a great impression on
me and are deeply appreciated.

I see myself as having been very fortunate in having as
supportive and as knowledgeable a guidance and dissertation
committee as I did. Drs. Harvey Clarizio, William Mehrens,
Louise Sause, and Robert Zucker have all been extremely help-
ful not only in regard to their input in this study, but also
in their overall influence on my professional competencies.
This influence has occurred through the courses they taught
and then later through the wise and thoughtful comments and
suggestions they provided while I planned my program of study.
Each of these people has been generous enough to share with
me his/her areas of expertise and has demonstrated competence
worthy of respect and emulation.

In addition to these faculty members at Michigan State
University, the staff at the Developmental Assessment Clinic
at Sparrow General Hospital was instrumental in helping to
get this project "off the ground". The task of data collec-
tion was made considerably easier by these individuals who
provided unlimited time and assistance. Dr. Eugene Dolanski,
Dr. Padamani Karna, John Wallen, Nels Bullock, and Joyce

Van Voorhis were all vital parts of this process. In addition,

V

Dr. Jerrold Veldman and his associates were especially help-
ful at crucial points in the data collection process and
their good natured assistance and cooperation were heartily
appreciated.

I owe much to Al Hammer of the M.S.U. Office of Research
Consultation who spent a considerable amount of time working
with me in analyzing the data and making efficient use of the
computers. His skills and suggestions proved to be highly
valuable.

I would also like to thank Donna Hobart for the excel-
lent job she did in keypunching the data, and Susie Breck for
her expert typing of this manuscript in its final stages.

I would be remiss indeed if I failed to acknowledge the
good natured cooperation of the mothers who took part in this
study. Their participation meant sacrificing much extra time,
and this clearly is a luxury new mothers have little of!

Finally, I want to thank my friends and "families" whose
love, support, and assistance have contributed directly and
indirectly to all that I have accomplished over these past
years at M.S.U. In particular, I want to thank Mindy, whose
love, understanding, and sense of humor brightened up the
grayest Michigan days and who constantly supported me even

though she knew "it wasn't going to be easy".

vi

TABLE OF CONTENTS

LIST OF TABLES . . . . . . . . . . .

Chapter
I.

II.

III.

IV.

INTRODUCTION . . . . . . .

Need for the Study . . .
Purpose of the Study . .

THEORY AND RESEARCH . . . .

Review of Theory and Research

Summary and Implications

NE THOD I I I I I I I I I I

Subjects . . . . .
Instruments . . . .
Testable Hypotheses
Analysis . . . . .
Procedures . . . .

RESLII'TS I I I I I I I I I I

Findings . . . . . .

Supplementary Analyses .
Summary . . . . . . . . .

DISCUSSION . . . . . .V. .

Results I I I I I I I I I
Practical Issues . . . .
Methodological Issues and

of the Study . . . . .
Future Research . . . . .
Implications . . . . . .
Summary and Conclusions .

vii

Page

ix

APPENDICES . . . .

A Nine Categories of Temperament and
Their Definitions . . . . . . . . . . .
B Data Sheet for Recording Information from
Mothers' Reports and Physician/Clinic
Records I I I I I I I I I I I I I I I I
C Consent to Participate Forms for
M0 thers I I I I I I I I I I I I I I I I
D Mother's General Impressions of Infant's

Temp erament (Carey/McDevi tt)

E Infant Temperament Rating - Psychologist's

Form

F Infant Temperament Questionnaire, Scoring
Sheet, and Profile Sheet

LIST OF REFERENCES

viii

Page
114

114

116

117

118

120

121
131

LIST OF TABLES
Table Page

5.1 Profile Sheet with Carey & McDevitt's Means
and Standard Deviations 50

5.2 Diagnostic Clusters for Individual Scoring
of Infant Temperament Questionnaire 50

5.3 Means and Standard Deviations for the Nine
Temperament Categories in Average and
Biological Risk Populations 51

5.4 Frequency Distribution - Temperament by
Biological Risk Status 52

5.5 Temperament Distributions in the Carey-
McDevitt Standardization Sample, the New
York Longitudinal Study and the Saltzman

Study (in percentages) 52
5.6 ANOVA Table for Total Self-Concept over

Temperament Category and Biological Risk 54

Status

5.7 Multiple Classification Analysis for Total
Self—Concept over Temperament Category and
Biological Risk Status 54

5.8 Frequency Distribution - Mother's Ratings
of Temperament by Psychologist's Ratings 56

5.9 Frequency Distribution — Temperament (as
measured by the I.T.Q.) by Mother's General
Rating 57

5.10 Frequency Distribution - Temperament (as
measured by the I.T.Q.) by Psychologist's
General Rating 59

5.11 Frequency Distribution - Temperament (as

measured by the I.T.Q.) by the Denver
Developmental Screening Test 60

ix

Table Page

5.12 Frequency Distribution of Responses to the
Question: "Do you feel that your child has
influenced the way you feel about yourself?" -
in Different Temperament Categories 63

5.13 Frequency Distribution of Responses to the
Question: "If response to the above is yes,
how has your child affected you?" - in
Different Temperament Categories 63

5.14 ANOVA Table for Family Self-Concept over
Temperament Category and Biological Risk
Status 66

5.15 Multiple Classification Analysis for Family
Self-Concept over Temperament Category and
Biological Risk Status 66

5.16 ANOVA Table for Physical Self-Concept over
Temperament Category and Biological Risk
Status 68

5.17 Multiple Classification Analysis for Physical
Self-Concept over Temperament Category and
Biological Risk Status 68

5.18 Frequency Distribution - D.D.S.T. Score by
At-Risk Status 70

INTRODUCTION

Need fnr the Study

Over recent years increased attention has been devoted
to the reciprocal and mutually reinforcing interaction be-
tween infants and their mothers. Not only has much research
been concerned with assessing parental attitudes and their
influence on the child's personality development (Medinnus
and Curtis, 1963), but there has been increased investiga-
tion of the child's contribution to the parent-child inter-
action in terms of "stimulating as well as being stimulated"
(Bell, 1971).

Furthermore, there has been much study of the way in-
fants differ from each other. In addition to obvious ways
such as size, rate of growth, and eye color, an extremely
important factor which is less obvious has also been studied.
"Temperament", as this factor is called, basically refers to
the usual manner in which infants react to other people and
things around them. There is no concrete evidence as to
whether temperament characteristics are inherited, whether
they are formed when the baby is growing during pregnancy, or
whether they develop in the first weeks of life. Berger
(1973) notes that there is likely an interaction of several
factors including: genetic sources, differences in brain
growth, sex linked determinants, and prenatal and perinatal

1

2
factors, all of which determine individual differences.
Scholom (1975), after reviewing the research in this area,
observed that temperamental attributes had been shown to be
present at birth, remaining constant to varying degrees over
time, and have a genetic component. In any case, temperamen-
tal qualities have been found in the first months of life and
there is a growing belief that understanding of the effect of
these on the child and his interactions is quite important.
As such, there has been much interest in how the temperament
of the infant interacts with the environment in molding the
personality of the child (Thomas, Chess and Birch, 1963).
These researchers, in particular, have shown that how a mother
interacts with her child depends on her personality, her
maternal feelings, and her infant's unique temperament.

The attention to the concept of self over the past years
has made it an important means of studying and understanding
human behavior. The individual's concept of himself has been
demonstrated to be highly influential in much of his behavior,
and directly related to his general personality and state of
mental health (Fitts, 1969). The research of a number of
personality theorists has suggested that the attitudes held
toward the self are quite important and are reflected in the
attitudes toward others. Fromm (1939), for example, has ob-
served that self love and the love of others go hand in hand.
It has also been noted that an individual develops his self-
definition from others, and it is generally agreed that the

child's self-concept is foremost an outgrowth of his home and

3

the people who live with him. Parents are extremely impor-
tant in shaping the self-concept of their children (Cooper-
smith, 1967). Gordon (1959) suggests that the child's origi-
nal self-concepts are the result of his interactions with his
parents and the particular meanings he assigns to these exper-
iences. Research indicates that low self-esteem in the
mother is associated with low self-esteem in the child (Tocco
and Bridges, 1973). Turner (1968) notes that "each person's
self-conception is a selective working compromise between
his ideals and the images forced upon him by his imperfect
behavior in actual situations" (p. 9h). It would seem that
the self-concept of the mother could thus be influenced by
her "imperfect behavior in actual situations" with her child.

In addition to these areas just mentioned, a recent
development has been the increased impetus to locate and then
to provide for children with special education needs. "Child
Find" - as programs with these goals have been called - is
both a state and a national commitment to locate and to
assess persons who have special needs from birth through
seven years of age. The provision of programs and services
to assist parents in the development and training of infants,
youth, and children, has been a common practice throughout
our country within our educational delivery system. The
concept of making such services available to parents of indi-
viduals who are impaired or who appear to have difficulty in
developing in either the cognitive, affective, or psychomotor

domains, is part of this increased commitment. Educators

4

today are increasingly concerned with early intervention
with children who are suspected of being delayed in one or
more of the three previously mentioned domains, this espe—
cially in terms of providing educational resources to assist
their parents. These resources would include: inservice
training, supportive technical services and counseling, that
would occur both in the home and in a central location.
There is the belief that in considering the provision of
services to the "high risk" 0 - 7 year old population, it is
essential that the total individual be considered.

It is the opinion of this writer that the initial areas
that have just been discussed, namely the mother/child inter-
action, infant temperament, and the issue of self-concept-—
specifically related to the mother, are all important points
to consider with regard to the last subject discussed, which
was the topic of "high risk" screening. More information
about the interaction of infant temperament and the self-con-
cept of the primary mothering adult in the child's life -
especially in this population of infants singled out for
possible developmental difficulties - would seem to be quite

important.
Purpose of the Study

After one year of encountering parents and infants in a
Developmental Assessment Clinic - which had as its purpose
assessing the development of babies who have received inten-
sive care during the first days or weeks of life - it appeared

that there were an overrepresentative number of "difficult

5
infants" among those with whom I had contact. Reports of
irregular, intense, active babies abounded, as did reports of
mothers who saw themselves as "failures as mothers". While
this pattern was not completely dominant, it did appear that
a considerable number of such cases did exist, evidencing
this trend.

There were five basic purposes for conducting this re-
search project, which in part grew out of the above observa-
tions. The first was to examine infant temperament variables
in both "average" and "high risk" populations. A "high risk"
infant for the purposes of this study was one who has, or is
likely to develop, a handicap capable of interfering with his
normal growth and development and/or his capacity to learn.
In this study, in order to focus this broad, widely used term
of "high risk", I have equated the term with Tjossem's cate-
gories of "Biological/Established Risk". Tjossem has divided
the concept of risk status into three categories of vulner-
ability in infants which include: "biological", "established",
and "environmental" risk. He states that these three cate-
gories are not mutually exclusive; however, cases can fit
predominantly into one of the categories as opposed to the
others. "Environmental risk" is a term used to discuss in-
fants who are biologically sound but for whom early life
experiences, such as maternal care and physical and social
stimulation, are so limited that without corrective interven-
tion there is a high probability for delayed development.

This category includes infants who are living in poverty and

6

babies who are unwanted. "Established risk" is a term Tjossem
uses to described infants whose early, atypical development
is related to diagnosed medical disorders of known etiology
with well known expectancies for developmental outcome "with-
in specified ranges of developmental delay". An example of
this is an infant with Down's Syndrome. The final category
is termed "biological risk" and it specifies infants who
present a history of prenatal, perinatal, neonatal and early
development events suggestive of biological insults to the
developing infant. These events increase the probability of
later appearing problems in development. Thus, the last two
categories, "Established/Biological", will herein be referred
to when the term "high risk" is used, with particular empha—
sis on the biological component involvement.

"Average" infants in this study are those babies for
whom the likelihood of developing handicaps capable of inter-
fering with normal growth, development, and/or capacity to
learn do not exist. Questions to be explored regarding this
first purpose include the whole idea of whether pre-, peri-,
or postnatal difficulties found in "high risk" infants are
related to differences in temperament. This relates to the
questions discussed in the introduction of this work, dealing
with the concern of whether the evaluation of the temperament
of the "high risk" infant whould be made an integral part of
the high risk screening procedures currently underway in
Child Find Projects in Michigan and other states.

A second purpose was to examine the relation between

infant temperament and maternal self-concept. An effort was

7
made to determine whether mothers of "difficult children"
have lower self-concepts and greater feelings of inadequacy,
specifically in their role as mothers. In this project, I
have restricted myself to demonstrating an association between
maternal self-concept and temperament variables, rather than
cause and effect.

A third purpose was to determine whether there is a
relationship between the self-concept of mothers and their
infants' "biological/established" risk problems. That is to
say, "high risk" status was examined in relation to the self-
concept of the mother. For example, does a mother whose in-
fant has a "high risk" of developing a handicap capable of
interfering with his normal growth and development necessarily
feel less adequate and have a poor self-concept?

A fourth purpose was to examine the relationship between
the mother's self-concept and the way she perceives her in-
fant. That is, did mothers with low self-concepts tend to
see their infants as having more difficult temperament styles
than was indicated by more objective measurements? Did
mothers with high self-concepts tend to see their infants as
having less difficult temperament styles than was indicated
by these measures?

The fifth purpose was to attempt to examine temperament
variables in both "average" and "high risk" populations in re-
gard to performance on a developmental screening measure, sex
of the child, and ordinal position in the family. This inves-
tigation sought to determine whether or not there was a rela-

tionship between these variables and infant temperament styles.

THEORY AND RESEARCH

Review of Theory and Research

Differences among individual children were recognized
as early in history as the early Greek and Roman periods.
Thereafter through the centuries that followed these differ-
ences were intermittently acknowledged and repudiated, with
varying degrees of importance applied to individual varia-
tions (Bell, 1977). In the twentieth century the impetus for
studying individual differences has come from various sources.
Predominant among these has been the work of Arnold Gesell,
who along with recognizing the importance of individual
variation also conducted much empirical work supporting a
picture of "considerable latent power residing in the child"
(Bell, 1977). In addition, work by such theorists as Piaget
provided for increased recognition of the considerable capa-
bility and organizing power of the infant. Korner (1970),
in reviewing the literature, noted that a large part of the
study of individual differences in infants, and the infant's
contribution to the "reciprocal exchange" with the mother,
came from the field of psychoanalysis. Citing Freud as an
example, she noted his firm conviction that "each individual
ego is endowed from the beginning with its own peculiar dis-
positions and tendencies" (Freud, 1937). Korner reports that
early psychoanalytic studies stressing the importance of pre-
dispositions in the development of neuroses were the precur-
sors to studies of systematic observations of innate differ-
ences among infants (Korner, 1970).

8

9

Bell (1977), noting the extensive research attention to
infancy in the 1960's and the facts on early perceptual and
cognitive capabilities that have been provided, has observed
that "approaches to socialization that ignored these (capa—
bilities) had been barren of results". It has become more
and more apparent that infants have certain built-in emergent
structures with which they react to and affect the things and
people around them. There is evidence to suggest that the
human infant at birth "possesses well organized behaviors,
the ability to discriminate among stimuli, and to show a pre-
ference for some of these stimuli" (Rheingold, 1967). In
addition, there is evidence to indicate that the infant can
actively organize, integrate, and adapt to experience (Rhein—
gold, 1969). Furthermore, research is showing that children
are born with certain personality predispositions which very
likely affect the behavior and attitudes of their parents
(Thomas et al.,1963; Moss et aL, 1968; Bell, 1974). From the
earliest weeks of life, it is the infant who effectively pro-
motes many interactions between himself and his caretaker.
For example, Moss and Robson (1968) note that at one month of
age it is the infant who initiates roughly four out of five
interactions. It has been shown that the infant, through his
discriminative crying, looking, and smiling, facilitates and
maintains interaction (Bell, 1974). His influence shows not
only immediate effects but long term ones as well. Bell and
Harper (1977) have observed that "We expect parents to in-

fluence their children. People who live together usually

10
affect each other, but what is often overlooked is the extent
to which children influence their parents and other adults"
(pg. ix). French (1977) notes that even "a moment's reflec-
tion will tell any adult who has had more than casual contact
with a small child, that the infant can exert definite and
sometimes very strong influences on the adult's behavior".
Rheingold (1969) goes so far as to suggest that "the infant
more powerfully socializes the adult into parenthood than the
parent socializes the infant". Bell (1977) has observed that
especially in the case with firstborns, the parents seemed
not only overwhelmed with their new responsibility but also
somewhat helpless and confused. "In many respects, they
appear to be much less powerful than the infant whose behav-
iors are remarkably well organized to produce a given result."
Research by Clarke-Stewart (1973), among others, supports
Rheingold's observation, specifically insofar as attachment
behavior is concerned. Yarrow (1965) noted that maternal
behaviors were influenced by the mother's preference for in-
fant behaviors, which again is indicative of the interactional
relationship existing between mother and child.

Among the early empirical studies of individual differ-
ences in the behavior of infants and children, a monumental
work upon which much of this current research is based is the
New York Longitudinal Study which was begun by Thomas, Chess,
Birch, Hertzig, and Korn (1963). Data on the behavioral
characteristics of a large sample of children were gathered

from infancy through adulthood. The study has determined

11
that the aspect of initial differences in infants that was
most likely to have pertinence for later development was what
they termed "temperament". Chess and Thomas (1972) noted
that they had used the term temperament as it had been em-
ployed by such theorists as Guilford and Cattell. The term
basically refers to the behavioral style of the child and
contains no inferences as to etiology. It is a "phenomeno-
logical term" which is used to describe the characteristic
tempo, energy expenditure, focus mood, and rhythmicity that
typify the behaviors of an individual child (Chess and Thomas,
1973). "Temperament refers only to the how, not the why of
behavior and implies neither immutability nor permanence"
(p- 9).

As in other studies (see, for example, Kagan and Moss,
1962), evidence was found in the New York Longitudinal Study
(N.Y.L.S.) suggesting the stability of certain temperamental
variables over time. Temperament identification in the in-
fants in this project was quite consistent for the first two
years of life. In follow up years, however, modification of
temperament scores occurred as interactions with the environ-
ment progressed. Kohlberg et al. (1972) note that there
appears to be continuity in a child's temperament although
not necessarily in his or her motivation. Temperament refers
to the manner in which a child moves, talks or thinks. Moti-
vation, on the other hand, refers to what the child actually
does or thinks. One example of consistent temperament would

be a child's activity level which does remain fairly stable -

12
a passive infant is likely to become a dependent adolescent
and adult (Kagan, 1977).

Outcomes for any individual child were determined not
only by parental functioning, or environmental influences, or
by temperament characteristics of the child alone, but also
by an interaction between temperament and environment. These
researchers identified a high risk temperament factor which
resulted in a significantly greater proportion of children
developing behavior disorders. Not all infants with this
pattern, however, did develop problems. Thus, there was felt
to be an interaction between biological and socialization
influence. The N.Y.L.S. findings suggest that the behavior
disorders that were observed did not develop as a direct
result of the persistence of an unusual temperament attribute
but rather that, "disorder resulted from some train of events
set in motion by the attribute or something associated with
it" (p. 114) (Rutter, 1976). It appears that this issue is
related to Sameroff and Chandler's (1975) concept of the
"continuum of caretaker casualty", which focuses on the fact
that the environment has a major role in minimizing or maxi-
mizing early developmental difficulties in infants. These
theorists describe a wide range of possible outcomes that
can result from environmental variation, and note that most
infants who suffer perinatal problems have proven to have
"normal outcomes". This point should be kept in.mind when
examining the N.Y.L.S.‘s findings regarding temperament. It

was found that twenty-three percent of the group who developed

13
behavior problems were "difficult" in temperament while only
four percent of the group without behavior problems were
"difficult". These "difficult" youngsters would appear to
be more vulnerable to the development of behavior disturbances
and seemed to present more severe symptoms when they became
disturbed. Thus, while a large number of "difficult" chil-
dren did not become disturbed, a significantly greater pro-
portion did as compared with those with "easier temperament
styles". Thomas and Chess (1977) note that it was possible
to trace the ontogenesis of the behavioral disturbances in
terms of the interaction between difficult temperament and
the child's environment.

Other studies have been carried out which lend further
credence to this aforementioned idea of interaction. One of
these was undertaken by Rutter (1964), who noted that "diffi-
cult children" who in general show a high percentage of be-
havior problems typically exhibited disturbed reactions if
the demands for socialization were presented inconsistently
and by an impatient adult. Graham et al. (1973) also found
that "adverse temperament characteristics" rendered the child
more vulnerable to the harmful effects of family discord and
other stress factors. In another study, Scholom (1975) found
evidence to suggest that parent temperament was quite impor—
tant as predictive of child adjustment. The mother's role
in particular stood out in this research project. Thus again,
while such factors as temperament might not lead directly to

behavior disorders, they do apparently render the child more

14
vulnerable to the effects of such things as stress factors,
parental discord, and parent temperament. Perhaps low self-
esteem in the mother might also be another one of these
factors?

In the New York Longitudinal Study, Thomas et al. (1963)
conducted content analysis of infant behavioral records and
found it possible to characterize the individual behavior of
each infant in terms of nine categories. These nine cate-
gories included: activity level, rhythmicity, approach-
withdrawal, adaptability, intensity of reaction, threshhold
of responsiveness, quality of mood, distractibility, and
attention span/persistence. Clusters of these temperament
attributes were found to be especially important when con-
sidering the reactivity style of individual children. Three
groupings were found to occur with high frequency and to
quote Thomas and Chess (1973), "it seemed sensible to call
them by everyday terms" (p. 88). These were the "easy child",
the "slow to warm-up child", and the "difficult child". It
is this last cluster that will be of special concern in this
research endeavor. The child who fits into this category is
irregular in biological functioning, negative in responding
to new stimuli, slow in adapting to change, negative in mood,
and responds frequently with high intensity in his reaction.
As such, this child will be difficult to feed, a chore to put
to sleep, hard to dress, and difficult to bathe -- to mention
a few aspects of his behavior. Thomas et a1. (1963) noted
that these infants "were variously characterized by their

mothers, the interviewers, and other members of the research

15
team (i.e., the N.Y.L.S. team) in terms of a series of pejor—
ative labels ranging from the expression "difficult children"
by the more sedate and formal of our colleagues to "mother
killers" by the more graphic and less inhibited" (p. 75).
This child makes extra demands for handling that are likely
to stimulate reactions of resentment, guilt or helplessness
in the mother. It would not seem unusual to expect to find
these feelings when a child shows behavior patterns - such
as those just mentioned - which quite clearly deviate from
the dominant cultural definition of what a "good" infant is.
Such children do not fit the stereotype of what is "optimal
behavior" for an infant. Justice and Justice (1976) speak
of this when they discuss "cultural scripts", which are the
culturally sanctioned messages and myths about how mothers
and infants "should" behave and feel. Infants are expected
to fit into the "bundle of joy" or "Gerber baby scripts".
These are myths that give rise to expectations that a baby
is a constant joy, "a clear cuddly creature who smiles loving-
ly at his mother and sleeps peacefully in her arms". This
myth is promoted by television commercials and advertisements
in women's magazines. These babies perpetuated through these
myths, clearly are not the "difficult infants"! It seems
that these "difficult infants" and the relationships they
have with their mothers are what Brazelton (1974) is speak-
ing of when he states that mothers may often have difficulty
"reaching their child". He notes that the "caring mother"

often feels something she has done has made her relationship

16
with her child "go wrong". This feeling of inadequacy, it
would seem, would be easily reinforced by a child having those
characteristics of the "difficult child". It doesn't seem
too hard to imagine how easily such an infant could augment
self-doubt in the primary caretaker. Commenting on this,
Thomas and Chess (1977) note that some children "have charac-
teristics which make it extremely difficult, or in some cases
virtually impossible for parents, no matter how good, to
function in such a way so as to guarantee an easy, smooth
developmental course" (p. 192). This is quite important to
keep in mind when noting a study carried out by Diggony,
Klein and Cohen (1964). They found that estimates of one's
capacities are highly dependent on the moment to moment feed-
back provided by the environment. Furthermore, these re-
searchers, as well as Kiesler and Baral (1970), noted that
actual performance of work is, in fact, affected by these
estimates. When it appears that they will fail, subjects
diminish their efforts. Not to believe that you are capable
of success may actually guarantee your failure. Perhaps this
might be the case with mothers of "difficult children". If
they see themselves as failures, will their efforts and
behavior indeed be hindered? This possibility is made more
plausible if one examines studies such as that carried out by
Robson and Moss (1970). These researchers discovered that
in some mothers "attachment decreased after the first month
if crying, fussing and other demands for caregiving did not

ebb". They note one report of a mother in just such a

17
situation who felt "strange and unloved, and wanted nothing
to do with her child" even though during pregnancy she was
"positive in orientation". Korner and Grabstein (1966) and
IMnn1(1977)also focus on this issue of maternal confidence.
They observed that early in an infant's life most common
interactions with the mother revolve around comforting and
soothing the infant. Infants were found to vary in their
need for comforting and in how soothable they were. The
impact of these differences on a mother - especially an
"inexperienced mother" - was thought to be "considerable".
Korner et al. (1966) state that the mother's feelings of self-
confidence as a caretaker and her feelings of relatedness to
her infant must be tied up very much with her effectiveness
in comforting her baby. Brazelton (1961) and Prechtl (1963)
also found that extreme irritability and inconsolability on
the infant's part can have "detrimental" effects on the early
mother—infant relationship. Bell and Ainsworth (1972) note
that some mothers and their infants get into vicious cycles
where the more the infant cried the more the mother withdrew
from efforts to soothe the child, particularly in the last
part of the first year of life. Once again in these cases
the stress is placed on the individual child's make—up -
not simply the Skill and devotion of the mother - as a strong
contributing factor to the mother's effectiveness. Thomas
and Chess (1965) observe that what looks like "bad mothering"
often shows up after careful investigation to be "the mother's

confused reaction to a difficult child, rather than a primary

18

cause of the child's problem". It seems to this writer that
negative effects such as this could relate particularly to
the mother's self-confidence.

Dodson (1970) addresses himself to the importance of
the self-concept of the mother in our society. He notes that
mothers are afraid to admit feelings of inadequacy regarding
their maternal role to anyone. "After all", he notes, "she
reasons she is a mother now and this is the crown and ful-
fillment of her life as a woman. She mentally pictures other
mothers quite adequately taking care of their young babies.
She thinks that she is the only one feeling this way so she'd
die before she would admit these feelings to anybody else"
(p. 32). Laing (1967) recognizes the awesome responsibility
mothers must feel, "They are always the first to get the
blame for everything". McBride (1973) notes that "since
women are supposed to be fulfilled by motherhood and in charge
of the family happiness, it seems to be entirely their fault
if they feel inadequate or if their children don't act right".
Considering our society's expectations, and keeping the
above points in mind, it seems understandable that the prob-
lems that might arise with child rearing could significantly
influence mothers. Dodson observes that new mothers receive
all kinds of messages from our culture, "some implicit and
some explicit", telling them that somehow mothers are equipped
with, as he puts it, "mother love and a mother instinct"
which automatically enables them to care for their babies.

This very often conflicts with mothers' actual feelings of

19

inadequacy in taking care of their infants. I would expect
this discrepancy to be even more extreme when the baby fits
into that category described as the "difficult child", and
the worrying and inadequacy feelings that result to be even
more extreme. No doubt the mother's feelings of helplessness
and self-doubt would clearly strain the mother/child rela-
tionship. Church (1976), too, emphasizes quite vehemently
the essential aspect of parental confidence in child rearing.
He notes that early in his life, the infant is able to sense
his parent's anxiety in dealing with him and is frightened
by this. Church notes that these feelings of inadequacy in
the mother and father are generally overcome "once the parents
have grasped that it is they who are in charge" (p. 20). Bell
and Harper (1977) attend to the fact that during the neonatal
period the mother learns that she can distract her infant
and thus begin to bring behavior under her control. "She
feels effective because the infant is doing what she wants
it to do, even though she may not understand that it is ceas-
ing the fussing for reasons other than her actions" (p. 134).
Mothers learn that not only are they under control but that
they have reinforcement value. One must wonder, however, if
this feeling of "being in charge" and "being effective" could
actually occur with a "difficult child". Is this lack of
confidence in the mother that Church speaks of, in fact,
overcome?

As has been indicated, Dodson recognizes the fact that

mothers who don't feel as though they have "a natural mother

20
instinct", think that all others do. Thus it seems, these
mothers are comparing themselves to other mothers. If a
mother "staked her all" - to paraphrase James - on being a
mother, and as such saw this as being a major role in her
life, it would naturally be quite important for her to do a
good job at it. To quote Hamachek (1971), "...it would be
difficult to maintain, not to mention enhance self-esteem
if one saw himself consistently finishing somewhere behind
the group with which he was comparing himself" (p. 9). A
mother rearing a "difficult infant" - and having a hard time
at doing it - might very likely see herself as a failure,
especially considering the fact that the maternal motive is
usually held up as our cultural ideal.

A special situation to examine when considering this
issue of maternal self-concept concerns "biological high risk
infants". Previously the effect on a parent of having a
"high risk" infant (specifically a premature baby) has been
examined in several studies (Caplan, 1960; Kaplan et al.,
1960; Caplan et al., 1965; Klaus et al., 1970). Bertrand
Cramer (1976) has observed that the most frequently found
feeling in mothers of premature infants was a sense of failure.
He further puts forth the idea that the birth of a premature
child is a severe blow to the mother's "self-esteem, her
mothering capabilities, and her feminine role". It is con-
ceived of as a loss of a body part, an insult to her body,
and a sign of inferiority. Cramer also noted that these

mothers expressed considerable feelings of guilt. They accused

21
themselves from the start of being bad mothers since they had
exposed their child to great stress "while forcing him out
of the protective womb". These feelings were especially
found to interfere with the establishment of relationships
with their children. The mothers felt as though they were
dangerous and incapable of protecting the child. The fact
that the premature infant is quickly taken away from the
mother following birth further serves to reinforce the belief
that the nurses can offer better maternal care than she can
and thus she is inadequate (Cramer, 1976). Klaus and Leider-
man (1977) found evidence to suggest that mothers of small,
premature babies who were separated from them in the first
few weeks of life (because the infants were in incubators or
in intensive care nurseries) tended to be less confident and
less attentive to their babies. They note that this might be
due in part "to the rather unresponsive nature of the very
premature infant" and in part to the lack of contact with
their infant during that early period "when mothering skills
should be developing". From this statement it can be seen
that the infant's behavior is also believed to influence the
situation. Drugs given to the premature infant, or anoxia
(which premature infants are often predisposed to), are two
causes frequently focused upon as explanations for variations
in the behavior of premature babies. Justice and Justice
(1976) note that anoxia frequently causes increased irrita-
bility and fussiness and that drugs often influence alertness

and thus responsiveness. It would seem that the premature

22
birth, the unfinished pregnancy, the early separation between
mother and infant, and the infant's behavior could all in-
fluence the mother-child relationship.

Prugh (1953) put forward the idea that the stress of
prematurity itself could very well "color the behavior and
attitudes of the infant through many years of childhood".
Caplan looked at the situation of having a premature infant
as a crisis situation and explored "healthy resolutions" to
the "crisis". Kaplan and Mason (1960) have isolated four
major psychological tasks which they felt were related to the
mother's experience of the prematurity crisis. They state
that the successful resolution of these tasks is essential
for the development of a healthy mother-child relationship.
One of the tasks that is paramount to accomplishing other
goals, is the need they place on the mother to acknowledge
her "feelings of failure" due to having not delivered a
"normal, full term baby". It would seem that the accomplish-
ment of this goal - and others - would be made especially
difficult when the child had those temperament characteris-
tics which were previously noted when discussing the "diffi-
cult child". To have added to the prematurity crisis an in-
fant who makes child rearing "extremely difficult" would, it
seems, very likely compound the feelings of failure, lower
self-esteem, and make the resolution of the crisis very
difficult and all the more painful for the mother. Caplan
et al. (1960) assert that mothers who deliver before term

have frequently been observed to urge their infants to eat

23
and grow as if to reassure themselves about their abilities
as mothers. They state that if this emphasis on the satis-
faction of the mother's own needs outweighs her concern for
the needs of the infant "a mother-child relationship may be
established which will interfere with the healthy personality
development of the child". Such a relationship is likely to
exist in a situation where a mother's self-concept is suffi—
ciently poor enough to hinder her full effectiveness as a
parent. Furthermore, as will be discussed shortly, if such
parents are in specific need of nurturing themselves, they
may be especially vulnerable to abusing their children, if
the children are not meeting these needs (Justice and Justice,
1976).

While this discussion of "high risk" has focused speci-
fically on prematurity, Cramer (1976) also notes that the
blow to the mother's self-esteem is likely to be even more
dramatic when her child is afflicted with a congenital defect
or a chronic disease. If parents of children born with mul-
tiple congenital defects do not receive help, Cramer notes
that they "can have such a lowered.self-esteem that they
totally withdraw interest in, and concern for, the child whom
they perceive as a sign of their own badness". Cummings et
al. (1966) state that variations in the physical, intellectual
and psychological characteristics of children likely influence
the parents' own perceptions of worth. They have found
evidence suggesting that having a child who is in some way
"deficient" is a highly stressful experience for the mother.

Cramer feels that the results in these situations can be very

24
"dramatic" with total maternal deprivation or "battered
child syndrome" as the outcome. As Cramer puts it "the
establishment of motheringbehavior is a fragile process"
(p. 164).

If there is a relationship between "difficult tempera-
ment" in infants and low maternal self-esteem, a logical
question to ask would appear to be, "What is important about
the self-esteem of the mother?" There have been many studies
in the past that have examined the relationship between the
mother's self-concept and issues related to her child's
development and her relationship to him. Sears, Macoby and
Levin (1957) found evidence to suggest that the mother's self-
esteem is an important correlate of her ability to express
warmth towards him. High aggressiveness in the child was
noted to be related to factors in the mother including:
anxiety about whether she was going a good job, low self—
esteem, and low value for the maternal role (i.e., she didn't
feel her primary responsibility should be towards her chil-
dren). Low self-esteem was also found in mothers who were
inconsistent in their relations with their children. Cooper—
smith (1967) notes that mothers who had high self-esteem had
a greater likelihood of accepting their roles as mothers and
carrying them out in a realistic and effective manner.
Samuels (1977) states that maternal childrearing attitudes
are related to how a mother feels about herself. "A mother
who feels inadequate in the mothering role will communicate

these feelings to her child by her verbal and nonverbal

25

behavior" (p. 88). An important point that has been reported
in the literature on child development is that mothers' self-
concepts are related to the self-concepts of their children
(Malone, 1967; Tocco and Bridges, 1973). Tocco et al. (1973)
note that self-concept measures have been found to relate to
many facets of life and the importance of parents in shaping
the self-concepts of their children has been found to be quite
significant. Studies by Combs and Snygg (1959) and Bayley and
Schaefer (1967) also address themselves to this very issue.
There is, in turn, much research existing which links the
child's self-concept to various areas in his life including
school achievement (Purkey, 1967) and numerous personal and
social variables (Fitts and Hammer, 1969). A child who sees
himself with confidence is likely to succeed just because of
the effect of this attitude on the way he Sets about the task.
After considering the ideas and studies that have just been
noted as well as other related research studies, it seems
'possible to reason that perhaps infants and their mothers
mutually influence each others' self-concepts. Since it has
been noted that parent-child interaction occurs in a recipro-
cal social system, it appears plausible to assume that each
contributes to the way the other feels about himself. This
is an idea that would seem to warrant further exploration.

In addition to maternal self-concept, there are other
areas that merit further investigation in relation to infant
temperament variables. Other variables that have previously

been examined in research studies have concentrated on the

26
sex of the infant, the I.Q. level, the ordinal position of
the infant in the family, and the temperament of infants "at
biological risk". With regard to this last point, high risk
status, minimal research has been carried out in the past,
insofar as its relation to temperament. In one study examin-
ing psychiatric disorders in children with congenital rubella,
Chess, Korn and Fernandez (1971) noted that "damaged young-
sters" (i.e., youngsters with physical defects) were more
frequently found in the "difficult child" category and
"undamaged youngsters" more frequently found to be "easier"
in temperament than random distribution of these characteris-
tics would lead one to expect. Clearly, there is a need to
examine relationships such as these in greater detail, some-
thing this current study has sought to do.

Regarding temperament and I.Q., Sameroff (1974), using
the Carey Infant Temperament Questionnaire and the Bayley
Infant Scales found that "if one wants to predict an infant's
I.Q. score at thirty months of age from a child's behavior at
four months of age, a much more reliable prediction can be
made based on his temperament than on his intellectual
functioning" (p. 142). Thomas and Chess (1977) note that
their own data from the New York Longitudinal Study have not
been analyzed to determine correlations between performance
on these types of tests and temperamental variables. Sameroff
and Kelly (in an unpublished manuscript noted by Thomas and
Chess, 1977) examined infant temperament using the Carey

Questionnaire and related scores on this instrument to the

27
sex of the child, birth order, and several other variables.
No sex differences were found with regard to temperament in
their sample: however, firstborn children were rated as more
active and more adaptable than later—borns. It would appear
that these variables and their relationships are far from
conclusive and warrant further study, this particularly
insofar as they occur in the populations under study in this
current project.

The issue of temperament differences in children has
been increasingly noted in the literature in relation to
child rearing practices. Specifically focused upon has been
the need to increase mothers' sensitivities to their children.
Carey (1970), in fact, had hoped in developing his Infant
Temperament Questionnaire, that through its use mothers
might be directed to make specific observations of their
children and ultimately be better parents. This could occur
by helping parents to respond appropriately to specific in-
fant behaviors. Ainsworth (1964), too, has observed that
there is a need for focusing on the individual infant and his
particular way of behaving. She speaks of the need for "a
new kind of child development counselor". Rather than giving
general advice or instruction - which she notes are likely
to be ineffectual - the counselor must tailor any guidance
to the particular "motherschild dyad". Ainsworth states that
perhaps the best advice that can be given as a guide to
mothers is for them to gear their interventions to their in—

fants' signals. Korner (1971), in noting that infants differ

.28
significantly from each other at birth, suggests that there
is no one best way for providing good child care for all in-
fants. She states that, "the only way to do so is to respond
flexibly to the individual requirements of each and every
child" (p. 51). Korner thus warns against seeing the mother
and the care and stimulation she provides as solely responsi-
ble for the normality or deviation of her child's development.
"This stance feeds into the illusion that with the 'correct
methods' and the 'right attitudes', we are in control of our
children's destiny" (p. 60). She also notes that it produces
much unwarranted guilt in the mother. "Mothers must be en-
couraged to be committed to the absorbing task of figuring
out the optimal dose and type of stimulation that is most
beneficial for their particular babies at any given point in
their development" (p. 60). It is the task of the parents,
according to Ainsworth (1964), to neither seek to make the
infant conform to a preconceived mold, nor employ rigid prac-
tices without regard for the rhythms, sensitivities, and pat-
terns of the particular infant. Thomas and Chess (1977) con-
tend that most parents of "difficult children" with behavior
disorders were able to respond positively to their child once
they learned that their child's temperament existed indepen-
dently of their own abilities and functioning, and that
specific management approaches were required. Basically,
this resulted from the reassurance that their patient efforts
would finally be rewarded by a change in adaptation by their

child, who would eventually function on a level "congenial

29

to their own value system". In studying their clinic sample,
Thomas and Chess observed that parent guidance in the cases
of difficult children with behavior problems was as success-
ful as in the cases of easy children. Margret Mahler (1963)
among others has noted that "child raising is not easily
achieved by the average mother in our culture". It was
asserted by Thomas and Chess (1977), however, that the recog-
nition of a child's specific qualities of individuality and
the adoption of those child care practices most appropriate
to him can significantly lighten the difficulties of child
rearing.

It is a tenet of this study that mothers will find it
easier to understand, take care of, teach, and enjoy their
children if they pay attention to their individual tempera-

ment styles.

Summany and Implications

From those studies and theories that have just been
examined, it can be seen that there has been much attention
in the past devoted to the study of individual differences in
infants. Recently, there has been an increased focus on the
individual differences in infant temperament variables, as
well as on the reciprocal relationship between mothers and
their infants. Bell (1977) summarized this last point when
he stated that "parent-child interaction occurs in a recipro-
cal social system in which much of the progress toward cul-
tural norms involves mutual adjustment and accommodation".

Much of the findings are inconclusive at this time and the

30

need for further research clearly exists. In addition, the
issue of infant temperament and it's relationship to maternal
self-concept appears to be a topic that has not been explored
thus far. This would seem to be a subject that could generate
many future questions and much future research in itself.

After considering the studies discussed in this section,
the question again arises as to whether the evaluation of
infant temperament should be made an integral part of the
screening procedures used in such programs as Child Find.
An area that is particularly relevant to this issue is the
problem of child abuse which is currently receiving much atten-
tion. Harrington (1972) reports that children contribute to
their abuse. "There is evidence that children who are
battered show disturbances of sleep and feeding, cry exces-
sively and respond poorly to attempts to comfort them"
(p. 68). This sounds very much like the "difficult child"
spoken of in this project. Harrington goes on to note that
how far they contribute to abuse is impossible to say.
"There is, however, a strong presumption that irritable
babies, hypersensitive babies, colicky and unresponsive
babies are especially vulnerable" (p. 68). According to a
study by Morse et al. (1970), fifteen out of twenty-five
battered children studied were considered "difficult" by
their abusing parents. Justice and Duncan (1975) note that
the literature implies that even "normal" parents may abuse
a particularly irritating or difficult child. A question

that would seem important to keep in mind is whether some

31

children, perhaps, are so difficult temperamentally that the
average parent may be moved to excessive measures of manage-
ment. This, too, seems to be an area in need of future research.
Steele and Pollack (1968). in studying the children of abusing
parents, note that from a very early point in infancy, the
child is expected to Show "exemplary behavior and a respect-
ful, submissive, thoughtful attitude toward the adult". Such
standards are implemented at an inappropriately early age and
are hardly what one could expect from the average infant, let
alone from one who has a "difficult temperament style".
Having an infant who doesn't respond "the way the book says"
and in accordance with the needs of a parent with a poor self-
concept might, it seems, provoke a parent to violence. Thus,
it appears that consideration of the "difficult child" and
recognition of a situation where extreme disharmony exists
between mother and child, would be very important in pre-
venting situations from reaching that stage where abuse could
occur. This would clearly be a concern for screening clinics.

As has been noted, parents of "difficult children" might
understandably feel threatened and anxious, quite possibly
because they believe the turmoil and difficulties of caring
for their child expose their inadequacies as parents. In our
society, considering the emphasis placed on the mother's role
and the emphasis on her responsibility for the infant's
healthy development, it would likely be the mother who is
most affected. She may believe that she is incompetent as a

caretaker or as Thomas and Chess note, she may blame the

32

infant and resent any extra demands he places on her. In
any case, the necessary patience and guidance needed on the
mother's part, especially in dealing with such a child, will
have a greater likelihood of being missing. Thomas and Chess
(1977) believe that these mothers will be more likely to
"pressure, appease, punish, or vacillate, all the time com-
municating a host of negative feelings to the infant, such
as hostility, impatience, or bewilderment" (p. 71). It would
seem plausible that this would lead to intensification of the
infant's negative moods and increased difficulties in adapta-
tion. Thus, a vicious cycle is established which researchers
such am; Thomas et al. believe leads to the development of
behavior disorders.

It has been suggested that guilt feelings and feelings
of inadequacy are possible results of interactions between
mothers and their difficult infants. This remains to be
examined in this study. Should it in fact be the case, it
might point the need for intervention with similar populations
in the future. It is important to note that the idea should
n93 be to label parents and infants who fall into a high risk
category, but rather to offer them support and guide them to
helpful services. These services could, in part, help mothers
(gain more positive feelings about themselves as mothers, and
gain a greater understanding of their relationships with
their infants. This brings up the issue of primary preven-
tion, which consists of heading off a problem before it occurs.

Unlike secondary prevention, it reduces not only the prevalence

33

of the problem but also the incidence--the number of new cases.
It would seem that keeping track of "difficult infants" and
checking on their development and well being (as well as the
well being of their parents) might be a wise strategy for

preventing serious problems at crucial times in the future.

METHOD

Subjects

The subjects in this study included seventy-five "bio-
logical high risk" infants and a control group of seventy-
five "average" infants. Once again, a "high risk" infant,
for the purposes of this study, is one who has, or is likely
to develop, a handicap capable of interfering with his normal
growth and development or capacity to learn. The disability
may have its origin in the prenatal, perinatal, or postnatal
period of the infant's life (Department of Health, Education
and Welfare, 1963). Among the conditions or complications
constituting "high risk" in infants are the following
(Waterman, 1969)!

a. Infant born to a mother who carries genetic factors
such as translocation chromosomes producing Downs
Syndrome, hemophilia and others.

b. Infant whose mother has had an acute virus infection
during pregnancy (especially within the first 2-3
months) such as German Measles, certain strains of
flu, toxoplasma, and others.

c. Premature or low birth weight infant (short gesta-
tional period or under 5 and 1/2 pounds).

d. Infant injured at birth resulting from difficult
labor or delivery.

e. Infant with a congenital anomaly.

f. Infant with RH or ABO incompatibility or marked
jaundice from other causes.

g. Infant with respiratory distress syndrome.

h. Post-mature infant (carried in utero beyond 42 weeks
gestation or an infant with a placenta which demon-
strates insufficiency).

The high risk infants frequently had more than one of
the conditions listed above. It was noted that eighty per-
cent of these babies were premature and seventy-six percent

had respiratory distress syndrome. Nine of the seventy-five

34

35
infants seen were post-mature. From four to eight infants
were found to have the other complications listed above in
various combinations. These high risk infants came from the
population of babies seen at a developmental assessment clinic
which is operated in coordination with the Child Find Project
in Michigan. These infants have been brought to the atten-
tion of the clinic either by the pediatrician, or by the
hospital where the infant was delivered. All of these babies
had received intensive care during the first days - and very
often weeks - of life.

The control group was made up of "average" infants con-
tacted through private physicians and "well-baby clinics".
This group, then, included infants for whom the development
of handicapping conditions capable of interfering with normal
growth and development or capacity to learn is not suspected.

For the purposes of this study the final subjects used
in both the average and high risk samples consisted of boys
and girls between four and twelve months of age (corrections
were made for prematurity).

The mean age in months for infants in the average group
was 7.89 (S.D. 2.13) and 7.21 (S.D. 2.17) for infants in the
high risk group. A two-tailed t-test revealed no significant
differences between the mean ages of infants in the two popu-
lations (F = 1.03, p:>.80). It should be noted that the ges-
tational age was used as the age of infants born prematurely
in an effort to get as accurate an age comparison as possible

with full term babies.

36

The mother-infant dyads that were part of this final
group all had the following common demographic characteris-
tics: intact family status, all the mothers were the natural
parents of the child, and middle socioeconomic status.
Socioeconomic status was roughly determined by several fac-
tors and this examiner sought basically to eliminate extremes.
Included in these factors was the chief source of family in-
come - which had to be from the father's employment. This
could be as the operator of a family-owned business or farm,
a public service worker, an independent professional, or a
white or upper level blue collar worker making approximately
over twelve thousand dollars per year. The education level
of the spouses was also considered, with both mothers and
fathers minimally having high school diplomas. Fifty-six
percent of the mothers and sixty-two percent of the fathers
were, in fact, college graduates. Subjective judgment was
used on the part of this investigator in considering the
above factors in combination with the "status of the job",
the number of members in the family, and the area where the
family lived, in an effort to get a relatively "middle class"
sample. This was done informally using ideas discussed by
Warner (1960) so as to roughly estimate socioeconomic status.
It should also be noted that all of the mothers included in
the final population were not employed outside of the home
at the time this study took place.

A point that seems important to note concerns the use

of mothers in this study as opposed to any primary parenting

37
adult. A recurring "complaint" that was voiced on the part
of the fathers of infants in this study was, "Why aren't I
being asked these questions?" It was encouraging to note
the enthusiasm generally seen on the part of fathers with
regard to the caring of their infants. With current trends
the way they are, situations are such that mothers are not
necessarily the primary caretakers. Due to complications
that result in the research - with regard to whom the find-
ings could be generalized, and so forth 4 this study was
limited to those mother-infant dyads where the mother was the
primary caretaker. Observations regarding those situations
where the mother is not the primary caretaker might be impor-
tant for future research, especially with regard to those
particular questions examined in this research study.

It was interesting to observe that no mother refused to
take part in the study. At its conclusion, however, twenty-
one mothers of "high risk" infants and thirteen mothers of
"average" infants had failed to mail back their questionnaires.
In addition to this, six mother-infant dyads from the "high
risk" population and nine dyads from the "average" population
were dropped from the study because they did not meet one or
more of the above mentioned criteria. Collection of the data
was terminated when the necessary forms were received from
seventy-five mother-infant dyads in each population who met

all of the stated conditions.

38
Instruments

Two standardized questionnaires, a standardized screening
instrument, and two informal rating forms were used as part
of this project. All the mothers in the study completed the
Carey Infant Temperament Scale-Revised 1977 Form (Carey and
McDevitt, 1977) (see item F in appendix). The Infant Tempera-
ment Questionnaire, a forced choice instrument, is composed
of 95 statements to which the mother responds on a seven point
scale (ranging from "almost never" to "almost always") to
descriptions of her child's behavior. The scale, originally
developed in 1968, was used to determine infant temperament
by yielding scores on the nine temperament dimensions found
in the New York Longitudinal Study (Thomas et al., 1968) (see
item A in appendix). The validity criteria for this instru-
ment included the degree to which test scores compared with
the following: 1) an interview with the mothers based on
techniques used in the New York Longitudinal Study and scored
by the N.Y.L.S. raters, 2) the N.Y.L.S. sample scores of
temperament, and 3) the mother's general impression of the
infant. Carey (1970) reported that in general the validity
of the questionnaire was supported by all three criteria.

The main purpose of the 1977 revision was to improve psycho-
metric characteristics of the original form. Test-retest
reliability for this new form was reported to range from

.66 - .81 for the nine temperament categories with a median
of .75, while internal consistency for the total instrument

was found to be .83.

39

Thomas and Chess (1977) note that on a practical level
the Infant Temperament Questionnaire has already demonstrated
its usefulness through the various research studies that have
been carried out using it. They acknowledge that further
research with the instrument is needed, and that later study
"should contribute to a future definition of the value of
this instrument".

From comments written at the end of the questionnaire
in this project, it seemed that several mothers found the
process involved to be long and tedious. "Too many of the
questions say the same thing", was a complaint voiced by one
mother. The instrument was reported by Carey to take approx-
imately twenty-five minutes to complete. This appeared to be
a conservative estimate judging from the comments made by
several mothers, who noted that it took them about an hour
to complete. One was led to believe that this was more a
reflection of the conscientiousness of the mothers taking
part in this study than an indication of problems understand-
ing the questions. In any case, it seemed that the general
consensus was that the process of completing the instrument
was an "informative experience".

The mothers were also given the Tennessee Self-Concept
Scale, which was developed by Fitts in 1955. This is a
thoroughly researched self report scale which was created
from a clinical perspective. It measures self-concept across
many sub-areas, providing both an overall self-esteem score

and a complex self-concept profile (Robinson and Shaver,

40

1973). The items on the scale fall into one of five general
categories: physical self, moral-ethical self, personal self,
family self, and social self. Each of these areas is then
broken up into statements of self identity, self acceptance,
and behavior. Specific attention was paid to the total

P score (which reflects the overall level of self-esteem),
row 1P score - Identity (which reflects the individual's
basic identity, "what he is as he sees himself"), Column D —
Family Self (which reflects feelings of adequacy, worth, and
value as a family member), and Column A - Physical Self
(reflecting one's view of his body, health, or physical
appearance). These scores were focused upon because of the
expected belief that they would be related to variables being
examined in this study.

The T.S.C.S. consists of 100 self-descriptive statements
which the subject rates as being like or unlike herself.
There are five response categories for each question, rang-
ing from completely true to completely false. The norms in
the manual are based on 626 people aged 12 through 68 years
old, and subsequent research has used a wide variety of
sample populations. Test-retest reliability was reported to
be quite high and correlations were noted with major self-
esteem inventories. The Tennessee Self-Concept Scale was
computer scored and cards keypunched by Counselor Recordings
and Tests at Vanderbilt University.

Fitts (1965) has indicated in the manual of the T.S.C.S.

that the scale generally takes from ten to twenty minutes to

41
complete. It appeared that the mothers, on the average,
took about thirty minutes to finish this scale. There were
a handful of complaints about the questionnaire being "irrele-
vant" but this too, like the Infant Questionnaire, was
generally well received.

In addition to these two questionnaires, the mother was
also asked to comment on her infant's temperament in general,
and specifically on the nine temperamental categories. This
form put together by Carey and McDevitt (1977) allowed the
mother to make broad, unspecific statements about her infant's
temperament, and any additional comments she chose.

The psychologist involved in the screening also filled
out a form, prepared by this writer, to determine his rating
of the infant's temperament. This was modeled after the Carey
and McDevitt form and like that form provided the psycholo-
gist with the opportunity to make general statements about
the temperament of the infant (see item E in appendix).

Unlike the mother's form, however, the psychologist was asked
to comment on only eight variables and the total temperament
estimate, instead of nine variables and the total. This was
due to the fact that rhythmicity was excluded from the psycho-
logist's form because of the limited time that he spent with
the infant.

The infants in both populations were administered the
Denver Developmental Screening Test to examine the develop-
mental progress they had made. This scale was developed in

1967 by Frankenburg and Dodds and has been used extensively

42
in clinical and research situations. Meier (1976) notes that
it has been subjected to extensive and intensive reliability
and validity studies. It has been demonstrated that the
ratings on the D.D.S.T. agree highly with the Bayley Infant
Scales and the Stanford Binet (Frankenburg et al., 1975). It
should be noted that the Denver is not an I.Q. test nor is it
a definite measure of future intellectual ability. It is,
on the contrary, a screening device used to "aid the health
provider in detecting potential problems" (Frankenburg et al.,
1975). The D.D.S.T. form is made up of 105 tasks written in
the range of accomplishments of children in the age span
from birth to six years old. The items are presented to the
child to see if he could pass tasks at the age "the majority
of infants do". The items are arranged in four sections
(personal/social, fine motor adaptive, language, and gross
motor performance) and developmental progress is evaluated as
normal, questionable, or abnormal. Again, it should be noted
that if a child does not perform a selected task, the examiner
is alerted to the fact that the child may have a problem and

further, more comprehensive evaluation is warranted.

Testable Hypotheses

The following is a list of null hypotheses that were
formulated and tested in this study:

1. There is no relationship between infant temperament
and the biological risk status of the child.

2a. There is no significant interaction effect on self-

concept between infant temperament and biological risk status

43

of the infant.

2b. There is no significant difference between the self-
concept of mothers of biological risk infants and the mothers
of average infants.

20. There is no significant difference in self-concept
between mothers of infants in different temperament categories.

3a. There is no relationship between the mother's per-
ceptions of the child's temperament and the psychologist's
perceptions of temperament.

3b. There is no relationship between the mother's general
perceptions of the child's temperament and.the temperament
scores on the Infant Temperament Questionnaire.

3c. There is no relationship between the psychologist's
general perceptions of the child's temperament and the tem-
perament scores on the Infant Temperament Questionnaire.

4a. There is no relationship between infant temperament
and scores on a developmental index.

4b. There is no relationship between infant temperament
and the sex of the child.

4c. There is no relationship between infant temperament

and the child's ordinal position in the family.

Analysis

A number of quantitative comparisons were carried out
between the variables in this study using the chi square
statistic and Analysis of Variance. The hypotheses were all
tested using portions of the Statistical Package for the

Social Sciences (SPSS) (Nie et al., 1975). All analyses

44

were conducted at the Michigan State University Computer
Center.

The chi square statistic was used to examine the rela-
tionship between temperament and 1) biological risk status,
2) sex of the child, 3) ordinal position, 4) developmental
level, 5) mother's perceptions of temperament, and 6) the
psychologist's perceptions of temperament. This statistic
was also used to investigate the relationship between the
mother's and psychologist's‘perceptions of the child's tem-
perament. The effect of temperament of the child and
"biological risk" status on the self-concept of the mother
was analyzed using a two-way ANOVA.

Supplementary analyses of the effect of temperament and
risk status on family self-concept and physical self-concept
were examined using two-way ANOVA. In addition, the chi
square statistic was used in the supplementary analyses
examining the relationship between perceptions of how the
infant influences his mother and temperament cluster, and the
relationship between risk status and developmental level.
The t-test was used to examine the difference in total self-
concept in mothers who reported that their child had in-
fluenced them positively and those who felt that they were
negatively influenced.

The phi statistic (in two by two tables) and Cramer's V
(in other cases) were reported as measures of the strength
of the relationship when chi square was used.

The Multiple Classifications Analysis was used to

45
determine the magnitude of the effect of factors in the
ANOVA designs.

Epocedures

The procedures for collecting the data for this study
were ostensibly the same for both populations being considered.
In the "high risk" and "average" groups the mothers, after
arriving for their prescheduled Developmental Assessment
Clinic or Well Baby Clinic appointments, were given an expla-
nation of the current study. At this time questions were
answered and parents reassured about the confidentiality that
would be enforced during and after the study. They were in-
formed that names would not be used at any time and that they
would be able to withdraw from the study without penalty
whenever they chose to do so. When the psychologist had
explained the nature of the study and the processes involved
to the satisfaction of the parents, they were given the op-
tion of participating and were requested to sign a "consent
to participate form" (see item C in appendix). The parents
were also offered an explanation of the findings of this
study at its conclusion. For this they were requested to
provide their addresses on a separate piece of paper.

If the parents agreed to take part in the study, they
were given two questionnaires to complete at home. The
Infant Temperament Questionnaire and the Tennessee Self-

Concept Scale were both briefly explained to the mothers and

no

then given to them to complete at their leisure.* For this
purpose a stamped, self-addressed envelope was provided.
This was done since it was felt that the time required to
complete both forms - believed to be about forty minutes -
was excessive under these circumstances.

The psychologist then completed the data sheet (see
item B in appendix) based on file records and the mother's
reports. Any discrepancies between the mother's statements
and records were examined and verified. The mothers were
then asked about their infant's temperament in general and
given the opportunity to speak about their relationship to
their child. The psychologist also completed a similar form
in which he too looked at the infant's temperament in a broad
sense, and noted any significant aspects of the mother-child
interaction that he observed. (It should be noted that two
master's level school psychologists were involved in the
evaluation in both populations of infants.) These two forms
enabled the comparison of scores derived from the relatively
detailed and objective behavioral observations on the Infant
Temperament Questionnaire, the mother's general impressions
of her infant, and the psychologist's observations of the
baby.

Finally, the infant was given the Denver Developmental
Screening Test. As has been noted, in this study - as in the
Developmental Assessment Clinic - the D.D.S.T. was used for
*Note: This frequently brought chuckles from most of the

mothers who readily indicated that "leisure time" was some-
thing they "knew little about"!

47
gross screening purposes. Most importantly, the administra-
tion of the scale was used as an opportunity to provide a
variety of situations for observing the infant and his style
of behavior.

While all of the above procedures were carried out in
the hospital outpatient clinic or pediatrician's office, six
of the infants in the "average" group were administered the

D.S.S.T. in the mother's home for her convenience.

RESULTS

Findings

In this chapter, the findings of the tests of the formally
stated hypotheses, as well as supplementary analyses will be
examined. Before doing so, however, it would seem important
at this time to elaborate on the scoring and interpretation
of the Infant Temperament Questionnaire.

Carey's definitions of diagnostic clusters of tempera-
ment were used to classify the infant into one of five cate-
gories: "Easy, Difficult, Slow-to-Warm-up, Intermediate High,
and Intermediate Low". These clusters are based on scores
on the nine temperament variables described in the New York
Longitudinal Study by Thomas et al. (1963). For determining
the child's diagnostic cluster, special attention is given
to six of these variables in particular: activity, rhythmi—
city, approach/withdrawal, adaptability, intensity, and mood.
The scores that the infant obtains are compared to Carey's
norms (see Table 5.1) and then fit into one of the diagnostic
clusters (see Table 5.2). For comparison's sake, the means
and standard deviations for the two populations in this study
can be found in Table 5.3.

Carey defines the diagnostic clusters used for indivi-
dual scoring as follows: (From "Infant Temperament Question-

naire Profile Sheet", Carey
and McDevitt, 1977.)

Easy - Scores greater than mean in no more than two of
difficult/easy categories (rhythmicity, approach, adaptability,
intensity and mood) and neither greater than one standard
deviation.

Difficult - Four or five scores greater than mean in

48

49

difficult/easy cate ories (rhythmicity, approach, adaptability,
intensity, and mood . These must include intensity and two
scores must be greater than one standard deviation.

Slow-to—warm-up - As defined above but, if either with-
drawal or slow adaptability is greater than one standard
deviation, activity may vary up to 4.68 and mood may vary
down to 2.47.

Intermediate - all others. Intermediate High — four or
five difficult/easy categories above the mean with one greater
than one standard deviation, or two or three above the mean
with two or three greater than one standard deviation.
Intermediate low - all other intermediates.

Hypothesisnl: There is no significant relationship
between infant temperament and the "biological risk
status" of the child.

Examination of the five diagnostic clusters of tempera-
ment in the two infant populations revealed no significant
relationship between temperament and biological risk status.
This hypothesis was not rejected at the p:<.05 level (chi
square = 3.95) (see Table 5.4). The larger percentage of
"difficult" and "intermediate high" temperament clusters in
infants in the high risk group and the lower percentage of
"easy" and "intermediate low" temperament clusters in this
same population were not significant. Here again, for com—
parison's sake, the percentage of occurrence of the five
diagnostic clusters in Carey's standardization sample and
the New York Longitudinal Study has been reported (see
Table 5.5).

Hypothesis 2a: There is no significant interaction

effect on the self-concept of the mother between infant

temperament and biological risk status of the child.

This hypothesis was not rejected at the p<:.05 level
(F = 1.15, df 4,140), indicating that risk status and tempera-

ment taken together did not significantly interact to influence

50

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

.pmmom
o>apowos sass sasoao .cocpns sod ozam
.Pmudm,
o>flpmmoc omcovsfi hasoam .Honpfis .nphnnm MMHQ
o>flpflmom wade .pmmom .mmm .cvzcn lhmmm
.mnopmsao ofipmoswmfin
onflmnsowpmosa
PnoemnoQEoe permsH mo wcfinoom Hmscfi>wocH pom whopmsau oapmoswmfla m.m manna
swan .npmfic non swan o>fipfimom saws .pnmom .Qmm .cpzsn 30H H
nwfln kno> hpo>
so.m mo.H om.m mH.m HS.N ma.H om.H mo.H mm.m .o.mH-
as.m mm.m mo.m Hm.m ms.m mo.m sm.~ om.m 03.: coo:
mm.s sm.m mm.m ms.m mH.s Ho.m mo.m mo.m om.a .o.mH+
.anom
30H .pmao Boa .noa 30H o>fivmwoc oncopcfl hasoam .nonpflz .npznnm swan
.smosss .possposo .pmaosoa ooos .msopsH .paoos spas\ma< .aspssm spa>npo<

 

 

 

 

 

 

 

 

 

 

mconvos>oo osmocopm ass memos m.ppa>ooo2 a house spas poocm manhood fl.m oanme

51

Table 5.3 Means and Standard Deviations for the Nine
Temperament Categories in Average
and Biological Risk Populations

Average Po ulation High Risk Population

 

 

N=75 (N=75)
Variable Mean S .D . Mean S .D .
Activity 4.36 .64 4.45 .69
Rhythmicity 2.49 .61 2.53 .59
Appro ach/

Withdrawal 2.68 .64 2.89 .73
Adaptability 2.38 .50 2.49 .60
Intensity 3.52 .67 3.66 .71
Mood 3.03 .54 3.22 .65
Persistence* 3.28 .64 3.43 .85
Distractability** 2.46 .45 2.65 .70
Threshold 3.95 .60 4.02 .66

 

Two-tailed t-tests
* Significant Difference at p<<.05 level
** Significant Difference at p< .01 level

Count

Row Pct
Col Pet

52

Table 5.4 Frequency Distribution - Temperament by

Temperament Cluster

Biological Risk Status

 

 

 

 

Total Pct Difficult Slow-to- Inter. Inter. Easy Row
Warm-up High Low Total
9 l 8 : 14 I 18 : 26 75
Average 12.0 ' 10.7 . 18.7 ' 24.0 . 34.7
Risk _ .9'9 .3 - _s.3' _'__ _e.§ , 12.0 ' 17.3
Status 14 ; 11 : 18 : 13 . 15 " 75
High 18.7 14.7 24.0 ' 17.3 ' 25.3
Risk 60.9 I 57.9 I 56.3 ' 41.9 I 42.2 50.0
903 I 703 I 1200 I 807 I 12.7
Column 23 19 32 31 45 150
Total 15.3 12.7 21.3 20.7 30.0 100.0
Chi Square = 95 _ . . . _
Cramer's V = .1624 df - 4 Significance - .41

Table 5.5 Temperament Distributions in the Carey-McDevitt
Standardization Sample, the New York

Study (in percentages)

Longitudinal Study and the Saltzman

 

 

Carey/McDevitt N.Y.L.S. Saltzmanfiv
(N=203) (N=141) (N=150
Temperament Cluster
Easy 42.4 40.0 30.0
Intermediate Low 31.0 - 20.7
Intermediate High 11.3 — 21.3
Slow-to-Warm-up 5-9 15.0 12.7
Difficult 9.4 10.0 15.3
_Undesignated - 35.0 -

 

 

53
the self-concept of the mother. (Self-concept here is meas-
ured by Total P on the Tennessee Self-Concept Scale, which
reflects the overall level of self-esteem.) Thus, the effect
of each variable is uniform across all the categories of the
other. Since there was no interaction effect, the outcome is
much clearer in looking at the significant effects of each
independent variable (see Table 5.6).

Hypnthesis 2b: There is no significant difference

between the self-concept of mothers of "high risk"

infants and the mothers of "average" infants.

The difference between the mean self-concept scores of
mothers of infants at biological risk and mothers of average
infants was found to be statistically significant. Hypothe-
sis 2b was rejected at the p<:.05 level (F = 3.77, df 1,140)
(see Table 5.6). Examining the Multiple Classification
Analysis (see Table 5.7) to determine the pattern of the risk
status' relationship to the self-concept of the mother, it
was noted that the self-concept of’mothers of high risk in-
fants varied in a downward direction while mothers of average
infants varied in the direction of higher self-concept scores.

Hypothegis 20: There is no significant difference in

the self-concept of mothers of infants in different

temperament categories.

The difference between the mean self-concept scores of
:mothers of infants in the five temperament categories was
found to be statistically significant. Thus, hypothesis 2c
was rejected at the p< .01 level (F = 8.54, df 4,140) (see
Table 5.6). The Multiple Classification Analysis (see

‘Table 5.7) indicates a pattern whereby lower self-concept

54

Table 5.6 ANOVA Table for Total Self-Concept over Tempera-
ment Category and Biological Risk Status

 

 

 

 

Significance
Source of Variation DF MS F of F
Temperament 4 13604.784 8.542 .001
Category
Biological Risk 1 6014.815 3.777 .050
Status
Interaction Effect 4 1842.274 1.157 .333
Residual 140 1592.673
Total 149 1999-355

 

Table 5.7 Multiple Classification Analysis for Total Self—
Concept over Temperament Category and
Biological Risk Status

*Grand Mean for Total Self-Concept of Mothers = 347.69

Variable and Category Deviation

 

 

Temperament Category

Difficult -15.30
Slow-to-Warm-up -24.17
Intermediate High —21.07
Intermediate Low +13.63
Easy +23.62
Biological Risk Status
High Risk -9.36
Average +9.36

Multiple R Squared = .227

 

55
scores were found in mothers of difficult, intermediate-high,
and slow-to-warm-up infants, while on the other hand, higher
scores were evident in mothers of easy and intermediate-low
temperament infants.

The multiple R squared was found to be equal to .23 (see
Table 5.7) and through this descriptive statistic it could be
determined that 23% of the variance in self-esteem in the
mothers was explained by the temperament category and the at—
risk status of the infant.

Hypothesis 3a: There is no relationship between the

mother's perceptions of the infant's temperament and

the psychologist's perceptions of temperament.

This hypothesis was rejected, as a significant relation-
ship was found between the mother's perceptions and the psy—
chologist's perceptions of the infant (chi square = 23-19.
p<:.001). As can be noted in Table 5.8, there was agreement
in the rating of the psychologist and mother in 40% of the
150 situations. Psychologists were more inclined to judge
infants as difficult than were mothers. In 53.3% of the
cases there was disagreement with the psychologist rating the
infant as more difficult, while in only 6.7 instances did the
mother rate the child as more difficult than the psychologist
did.

Hypothesis 3b: There is no relationship between the

mother's general perceptions of the child's temperament

and the temperament scores on the Infant Temperament

Questionnaire.

A significant relationship was also found between the

variables in hypothesis 3b (chi square = 25.79, p<:.01) (see
Table 5.9). In 43.4% of the total cases studied, there was

Table 5.8 Frequency Distribution - Mother's Ratings of

56

Temperament by Psychologist's Ratings

Psychologist's General

Temperament Rating

 

 

 

 

Count
Row Pct R
Col Pct Easy Average Difficult T
Tot Pet
19 ' 47 l 8
Easy 25.7 63.5 | 10.8
82.6 52.8 21.1
12.7 . 31.3 | 5.3
MOther'S Li' ' 36 I 25
General Average
Temperament 1?:E : 28:4 l 2:23
Rating 2.7 . 24.0 I 16.7
0 . 6- I. - -5- -
. . 0 a 54.5 g 45.5
Difficult O 6.7 13.2
0 I [400 I 3'3
Column 23 89 38
Total 15.3 59.3 25.3
Chi Square = 23'19 df = 4 Significance = .0001

Cramer's V = .2780

ow
otal

74
49-3
65
43.3
11

7.3

150
100.0

57

 

Cramer's V

 

 

 

Table 5.9 Frequency Distribution - Temperament (as measured
by the I.T.Q.) by Mother's General Rating
Count
Row Pct Temperament Cluster - I.T.Q. Rating
001 Pet
Tot Pct Difficult Slow—to Inter. Inter. Easy Row
Warm-up High Low Total
9 i 8 ' 7 i 17 I 33 74
Easy 12.2 ' 10.8 ' 9.5 . 23.0 ' 44.6
39.1 42.1 I 21.9 ' 54.8 ' 73.3 49.3
Mother's 6.0 I 5.3 4.7 11.3 , 22.0
Temperament - - - - - e 4 - - -' ------
Rating 13 ' 10 ' 19 , 12 ' 11 65
20.0 ' 15.4 29.2 . 18.5 ' 16.9
Average 56.5 1 52.6 I 59.4 38.7 ' 24.4 43.3
8.7 I 6.7 ' 12.7 ' 8.0 . 7.3
_ _. _. ._ - __ _... __' ______
1 u 1 | 6 I 2 : 1 11
Difficult 9.1 l 9.1 54.5 18.2 9.1
4.3 . 5.3 I 18.8 I 6.5 ' 2.2 7.3
07 I 07‘ 400 I 103 I .7
Column 23 19 32 31 45 150
Total 15.3 12.7 21.3 20.7 30.0 100.0
Chi Square = 25'79 df - 8 Significance = .0011

.2932

58
agreement between the mother's general impressions and her
ratings of the child on the Infant Questionnaire. It was
noted that in approximately 42% of the cases the mothers
rated their child as easier than the questionnaire would in-
dicate, while in only 13% did they rate the infant as more
difficult than the child apparently was.

Hypnthesisg3n: There is no relationship between the

psychologist's general perceptions of the child's

temperament and the temperament scores on the Infant

Temperament Questionnaire.

A significant relationship was found between the psycho—
logist's general perceptions of the infant's temperament and
the scores on the Infant Temperament Questionnaire at the
p.:.0001 level (chi square = 64.02). The psychologist agreed
with the Infant Temperament Questionnaire rating in approxi-
mately 55% of the total cases. When there was disagreement,
the psychologist was more inclined to have rated the child
as more difficult than the questionnaire indicated he was

(see Table 5.10).

Hypothesis 4a: There is no relationship between infant
temperament and scores on a developmental index.

Examining the five categories of temperament and the
performance on the Denver Developmental Scales, a significant
relationship was indicated (chi square = 20.03, p<:.05) (see
Table 5.11). Out of 149 cases considered (one mother was
not able to arrange a convenient time for testing her infant),
116 infants (77.9%) were found to be at developmental levels
within normal limits, 30 infants (20.1%) were within the

questionable range, and 3 infants (2%) showed delayed

59

Table 5.10 Frequency Distribution - Temperament (as measured
by the I.T.Q.) by Psychologist's General Rating

Count

Row Pct
Col Pct
Tot Pct

Easy

Psychologist's

Temperament
Rating

Average

Difficult

Column
Total

Chi Square
Cramer's V

Temperament Cluster - I.T.Q. Rating

 

 

 

 

Row
Total

23
15-3

89

59-3

38

25.3

150

100.0

Difficult Slow-to Inter. Inter. Easy
Warm-up High Low
1 ' 1 I 0 . 2 j 19
4.3 : 4.3 I 0 I 8.7 , 82.6
4.3 5.3 , 0 , 6.5 ‘ 42.2
07 ' 07' O ' 103 1207
._.___ _. .. -_ -_--.-_-
9 : 12 I 17 . 27 I 24
10.1 ' 13.5 I 19.1 I 30.3 . 27.0
39.1 63.2 I 53.1 . 87.1 53.3
6.0 . 8.0 11.3 , 18.0 ' 16.0
———— ———~l ——————— I---
13 I 6 15 ' 2 I 2
34.2 ' 15.8 ' 39.5 ' 5.3 5.3
I
56.5 I 31.6 I 46.9 I 6.5 4.4
8.7 | 4.0 . 10.0 I 1.3 I 1.3
23 19 32 31 45
15.3 12.7 21.3 20.7 30.0
= 64'02 or = 8 Significance = .0001

.4619

Table 5.11

Count

Col Pct
ROW'PCt
Tot Pct

Normal
Denver
Development
Test Score

Question-
able

Delayed

Column
Total

Chi Square

Cramer's V

60

Frequency Distribution - Temperament (as measured

by the I.T.Q.) by the Denver Developmental

Screening Test

Temperament Cluster - I.T.Q. Rating

 

 

 

 

Row
Total

116

77.9

30

20I1

149
100.0

Difficult Slow-to Inter. Inter. Easy
Warm-up High Low

17 : 10 : 21 I 30 : 38
73.9 , 52.6 | 67.7 ' 96.8 I 84.8
14.7 8.6 18.1 I 25.9 I 32.8
11.4 I 6.7 ' 14.1 . 20.1 25.5
.. - .. - _ .. _l- _ - - — _ ..l_ ._ -
6 I 8 ' 8 : 1. 7
26.1 I 42.1 ' 25.8 3.2 , 15.6
20.0 26.7 26.7 ' 3.3 23.3
4.0 I 5.4 I 5.4 I .7 I 4.7
— — -- - - - — - - —- — — — —.-—-
0 : 1 ' 2 ' 0 . 0
0 5.3 ' 6.5 ' 0 I 0
O I 33.3 I 66.7 I 0 I 0
O I 07' 103 | 0| 0
23 19 31 31 45
15.4 12.8 20.8 20.8 30.2

= 20'03 df = 8 Significance = .0102

.2592

61

development. The largest number of developmentally "normal"
infants were in the "easy" temperamental category (38 out of
116 normal infants). In the "delayed range" no difficult nor
easy infants were found: however, one slow-to-warm-up and two
intermediate high infants were noted. In the "questionable"
group, these two temperament groups (i.e., slow-to-warm-up

and intermediate high) were again highly represented, although
the five categories were fairly evenly distributed.

Hypothesis 4b: There is no relationship between
infant temperament and the sex of the child.

This hypothesis was not rejected, and as such there was
not found to be a significant relationship between tempera-
ment and the sex of the child (chi square = 4.19, p) .30).
Thus, despite the fact that there were more than twice as
many boys in the difficult category and more girls in the
slow-to-warm-up cluster, these facts were not statistically
significant.

Hypothesis 4c: There is no relationship between infant

temperament and the child's ordinal position in the

family.

The fact that this hypothesis was not rejected at the
p < .05 level (chi square = 7.08) led to the conclusion that
there was no relationship between temperament and the child's
birth order in the family. The fact that there were nearly
three times as many difficult infants who were firstborns
as opposed to later-borns was a trend not found to be sta-

tistically significant.

62

Supplementary Analyses

In addition to the formal hypotheses that have just been
noted, several additional points have been examined.

Mothers were interviewed to determine whether there was
a relationship between self-concept, temperament of the child,
the at-risk status of the child, and whether mothers thought
their infant had influenced the way they felt about themselves
(see item D in appendix). Seventy-three percent of the 150
mothers questioned felt that the child had, in fact, influ-
enced the way they felt about themselves. Forty mothers
stated that their child had not influenced their feelings
about themselves. The relationship between responses to this
question and temperament of the child was found to be signi-
ficant at the p<:.001 level (chi square = 19.09) (see
Table 5.12). There was a trend observed that mothers of easy
children were more inclined to acknowledge that their child
had influenced them, while mothers of the more "difficult
tenperament" infants (including difficult, slow-to-warm-up,
and intermediate high clusters) were inclined to see their
child as n93 having influenced them.

It was also found that of the mothers who had reported
that their child had affected them, 92.7% indicated that this
influence was in a positive direction. The relationship
between the temperament of the child and the way the mother
saw her child as affecting her (i.e., in a.p0sitive or nega-
tive way) was found to be significant at the p<:.001 level

(chi square = 20.08). Examining this relationship in

63

Table 5.12 Frequency Distribution of Responses to the Question:
"Do you feel that your child has influenced the
way you feel about yourself?" - in Different
Temperament Categories

 

 

 

 

Count
Col Pct Temperament Cluster - I.T.Q. Rating
ROW'Pct
Tot Pct Difficult Slow-to Inter. Inter. Easy Row
Warm-up High Low Total
I I
15 I 11 1 17 I 29 ' 38 110
Yes 65.2 . 57.9 ' 53.1 93.5 , 84.4
. 13.6 I 10.0 I 15.5 I 26.4 34.5 73.3
Has Child 10.0 ' 7.3 I 11.3 I 19.3 I 25.3
Influenced -—.— — —-.— — .-'- .. __,_ - _'- - -
You 8 I 8 I 15 ' 2 I 7 40
34.8 ' 42.1 46.9 ' 6.5 . 15.6
No 20.0 I 20.0 I 37.5 I 5.0 17.5 26.7
503 l 503 I 1000 ' 103' “’07
Column 23 19 32 31 45 150
Total 15.3 12.7 21.3 20.7 30.0 100.0
Chi Square = 19.09 _ . . . _
Cramer's V = .3568 df — 4 Significance - .0008

 

 

Table 5.13 Frequency Distribution of Responses to the Question:
"If response to the above is yes, how has your child

 

 

 

 

affected you?" - in Different Temperament Categories
Count
Col Pct Temperament Cluster - I.T.Q. Rating
Row Pct
Tot Pct Difficult Slow-to Inter. Inter. Easy Row
Warm-up High Low Total
j
10 I 11 : > 15 I 28. 38 102
Positively 66.7 ' 100.0 88.2 1 96.6 I100.0
9.8 I 10.8 ' 14.7 I 27.5I 3K.3 92.7
How _ .9: -i - £09 I -136 1.2.5.? “3.?
Influenced 5 | O ’- 2 I 1' O 8
. J
‘ 33.3 ' o 11.8 ' 3.4I 0
Negatively 62.5 I 0 I 25.0 I 12.5. 0 7.3
Column 15 11 17 29 38 110
Total 13.6 10.0 15.5 26.4 34.5 100.0

Chi Square = 20.08

Cramer's v = 4273 df = 4 Significance = .0005

69
Table 5.13, it can be seen that of the mothers who felt that
their child had influenced themselves in a negative way, the
greatest number had children who were "difficult" temperamen-
tally. Furthermore, there were no mothers of "easy" children
in this group of "negatively influenced mothers".

In addition to the above, it was noted that there was
not a significant relationship between the biological risk
status of the infant and whether or not the mother felt she
had been influenced by the child (chi square = .03, p2>.80).
Furthermore, examining the self-concept of mothers who felt
that their child had influenced them positively as opposed
to those who felt they were negatively influenced, a signi-
ficant difference was indicated. Using a t-test to examine
mean self-concept in the two groups of mothers, a one-tailed
test with F = 2.08, p = .05, indicated significance. Mothers
in the "positively influenced" group had a mean self-concept
score equal to 360.36 while mothers in the "negatively
influenced" group had a mean of 330.37. The mean self-con-
cept score for all mothers in the study was 307.69.

The self-concept variable was a factor also examined
beyond the formally stated hypotheses in this study. One
subtest on the Tennessee Self-Concept Scale which was analyzed
in addition to the overall self-concept was the Family Self—
Concept. This score reflected the mother's feelings of
adequacy, worth. and value as a.family member, and as such
it was believed that this would be especially affected by

those variables previously noted in reference to the total

65
self-concept, namely infant temperament and the risk status
of the child.

Referring to Table 5.10, it can be seen that there is
no interaction effect between the two independent variables
(temperament of the infant and risk status) over the depen-
dent variable of Family Self-Concept (F = .98, df 4,190,
p:>.#0). Testing the main effects did reveal significance.
The analysis gives reason to believe that the differences in
temperament made a difference in the family self-concept as
did the differences in at-risk status. This is to say that
there was a significant difference in the Family Self-Concept
scores of mothers of infants in the different temperament
categories at the p<:.01 level (F = 7.39, df n.140). There
was also a significant difference in the Family Self-Concept
scores of mothers of "high risk" infants and the mothers of
"average" infants at the p<:.05 level (F = 6.59, df 1.1h0).
Examining the Multiple Classification Analysis in Table 5.15.
it can be noted that scores on the Family Self-Concept vari-
able are below the mean of the total population in the mothers
of "difficult", "slow—to-warm-up", and "intermediate high"
temperament infants (all towards the difficult end of the
temperament continuum range). Family self-concept scores,
on the other hand, are above the mean in "easy" and "inter-
mediate low" temperament categories (both towards the easy
end of the continuum of temperament scores). The self-con-
cept scores can also be seen to vary in a "positive" direc-

tion in the "average" population while varying in a "negative"

66

Table 5.14 ANOVA Table for Family Self-Concept over Tempera-
ment Category and Biological Risk Status

 

 

Significance
Source of Variation DF MS F of F
Temperament 4 806.720 7.390 .001
Category
Biological Risk 1 718.913 6.586 .010
Status
Interaction Effect 4 107.449 .984 .420
Residual 140
Total 149

 

Table 5.15 Multiple Classification Analysis for Family Self-
Concept over Temperament Category and
Biological Risk Status

*Grand Mean of Family Self—Concept of Mothers = 69.59

 

 

Variable and Category Deviation
Temperament
Difficult -2.76
Slow-to-Warm-up -5.85
Intermediate High -5.68
Intermediate Low +2.57
Easy +6-15
Biological Risk
High Risk -2.89
Average +2.89

Multiple R Squared = .222

 

67
direction in the "high risk" group.

Comparing the analysis of the Family Self-Concept varia-
ble and the Total Self-Concept measure, one can note the
very similar results in both instances. In the Family Self-
Concept, however, the results are significant at a higher
level. After examining the Multiple R Squared statistic, it
can be determined that 22% of the variance in Family Self-
Concept is explained by the temperament category and the
at-risk status of the child.

An additional self-concept variable which was examined
was the Physical Self-Concept scale of the Tennessee Self-
Concept Scales. This score reflects the individual's view
of his body, health, and physical appearance. It was believed
that this score would be related to temperament and risk sta-
tus as well. It can be seen from Table 5.16 that there is
no interaction effect between temperament and risk status
over this variable of Physical Self-Concept (F = 2.22,
df 4,140, p:>.05). In addition, there was not found to be a
significant difference in Physical Self-Concept between
"average" and "high risk" populations (F = 2.41, df 1.140,

p >.10). There was, however, a significant difference in

the Physical Self-Concept in the mothers of infants in the
different temperament categories (F = 6.82, df 4,140, p<:.01).
After examining the Multiple Classification Analysis (see
Table 5.17) it can be noted that the Physical Self-Concept
scores are below the mean in the temperament categories
towards the "difficult" end of the temperament continuum and

above the mean towards the "easy" end.

68

Table 5.16 ANOVA Table for Physical Self-Concept over Tem-
perament Category and Biological Risk Status

 

 

Significance
Source of Variation DF MS F of F
Temperament 4 644.240 6.818 .001
Category
Biological Risk 1 228.124 2.414 .120
Status
Interaction Effect 4 209.689 2.219 .070
Residual 140 94.490
Total 149 115.366

 

Table 5.17 Multiple Classification Analysis for Physical
Self-Concept over Temperament Category and
Biological Risk Status

*Grand Mean of Physical Self-Concept of Mothers = 67.95

Variable and Category Deviation
Temperament
Difficult -3.67
Slow-to-Warm-up -4.12
Intermediate High -4.82
Intermediate Low +3.54
Easy +4.83
Biological Risk Status
High Risk -1.91
Average +1.91

Multiple R Squared = .182

 

69

A final issue that was addressed in analyzing the data,
was a consideration of the Denver Developmental Screening
Test and the relationship between performance on this test
and the at-risk status of the infant. As can be seen in
Table 5.18, there was a statistically significant relation-
ship between the scores on the Denver Developmental Screening
Test and the "biological risk status" of the infant at the
p<:.001 level (chi square = 17.24). It can be noted that
91.9% of the babies in the "average" population were at
developmentally "normal" levels while in the "high risk"
group, 64% were developmentally normal. No infants in the
"average" population were developmentally delayed, while
three babies were delayed in the "high risk" population. In
addition, out of thirty infants who were in the questionable
category, twenty-four of them were from the "high risk"

population.

Summary

Examining the analysis of the data in this chapter has
revealed several important points. Initially, it was found
that there was no statistically significant relationship
between temperament and "biological risk" status, sex of the
child, and the child's ordinal position in the family. There
was statistical significance, however, in the relationship
between the scores on the Denver Developmental Screening Test
and both the temperament of the infant and his "biological
risk" status.

The results also indicated that there was a significant

70

Table 5.18 Frequency Distribution - D.D.S.T. Score
by At-Risk Status

 

 

 

 

Count
Row Pct Denver Developmental Test Score
Col Pct Row
Tot Pct Normal Questionnable Delayed Total
68 i 6 : 0 74
Average 91.9 I 8.1 ' 0
58.6 20.0 0 49.7
Status -—-'-- —---'—___-
48 24 3 75
High 64.0 ' 32.0 ' 4.0
Risk 41.4 I 80.0 I 100.0 50.3
32.2 I 16.1 I 2.0
Column 116 30 3 149
Total 77.9 20.1 2.0 100.0

Chi Square = 17.24
Cramer's V = .3402

df = 2 Significance = .0002

71
difference in the self-concept of mothers of infants in dif-
ferent temperament categories. Mothers' self-concepts also
varied significantly depending on whether or not the child
was in the "average" or "biological risk" population. No
interaction effect was found between self-concept and risk
status.

Mothers' perceptions of their infants' temperament were
compared with psychologists' perceptions and scores on the
more objective Infant Temperament QuestiOnnaire. Significant
relationships were once again found, with mothers rating
their child as "easier" in temperament than both the ques-
tionnaire and the psychologist.

Supplementary analyses were also carried out examining
mothers' feelings as to whether they thought they had been
influenced by their child (and if so in what way). Signifi-
cant results were found indicating that most mothers thought
their child had influenced them in a positive way. Of those
who perceived their child as a negative influence, the great-
est number had children who were difficult temperamentally.
Furthermore, mothers who felt negatively influenced by their
child were found to have a significantly lower self-concept
than mothers who felt positively influenced.

Finally, Family Self-Concept and Physical Self-Concept
were examined in relation to temperament of the child and
biological risk status. Both the Family Self-Concept and the
Physical Self-Concept were found to differ significantly in

the different temperament categories. Only the Family

72
Self-Concept was noted to differ significantly between the
high risk and average groups.
Questions and issues generated by these results will be

explored in the following chapter.

DISCUSSION

In this chapter, in addition to examining and interpret-
ing the results that have been presented in the previous sec-
tion, there will also be a focus on: practical and methodo-
logical issues, limitations of the study, possibilities for
future research, implications of the study, and the conclu-

sions generated.

Results

The first, and most obvious point emerging out of this
study was a reaffirmation of the fact that infants, in no
uncertain terms, do differ widely in their individual tem-
perament styles. The distributions of temperament clusters
found in this project were similar to_the findings of both
the New York Longitudinal Study and Carey and McDevitt's
standardization of the Infant Temperament Questionnaire.

Earlier this writer reported several "armchair" analyses
made on the basis of many months of work as the psychologist
on a team screening "high risk" infants in a developmental
assessment clinic. As it turns out, the first of these be-
liefs, namely, that there was a relationship between "biolo-
gical risk" status of the infant and his/her temperament
style, was not proven to be valid statistically. While at a
descriptive level there were more "difficult infants" in the
"high risk" population and more "easy infants" in the "average"
group, statistical analysis indicated that the differences
between the two groups were not statistically significant.

73

74
It would appear that factors associated with "high risk"
infants in this study were not especially influential on the
temperamental characteristics of these babies. Related to
this, one might also assume that the label of being a "high
risk" infant did not appreciably influence the child's tem-
perament. Considering the current focus on "labels" and the
potential negative effects they are purported to have, this
point would seem to warrant further exploration. It should
be noted that no implication is made here that the lack of a
significant relationship between temperament and riSk status
will hold up over time. This is to say that a follow-up of
this "high risk" group would be desirable to determine whether
or not at some time in the future there might, in fact, be
indications of some type of influence.

Another "armchair" analysis did prove to be not only
descriptively significant, but statistically significant as
well. A relationship was found between the infant's tem-
perament and the mother's self-concept, as well as between
this same self-concept variable and the "biological risk"
status of the baby. Thus, mothers with lower self-concepts
were found more often in dyads with infants who had tempera-
ment clusters towards the "difficult" end of the continuum
(i.e., difficult, slow-to-warm-up, and intermediate high
temperament categories). Furthermore, mothers with lower
self-concepts were also found more frequently to have in-
fants in the "high risk" population. The implications of

these findings, while incomplete, are vast and complex, and

75
most definitely remain to be studied further. In any case,
several possibilities could be explored to explain these
findings. The first feasible idea that comes to mind is
that perhaps these "low self-concept" mothers do believe the
"conventional middle class wisdom" of our society which
states that they are solely, by their maternal attitudes and
motivations, influencing the course of their child's develop-
ment. Thus, with this in mind, easy children would seem to
reassure their mothers that they are adequate and loving
parents. These babies provide their mothers with ample oppor-
tunities to prove themselves and possibly even feel superior
to those mothers struggling with their "difficult infants".
The mothers with "difficult" babies probably are given fewer
opportunities to prove themselves as adequate. Where self-
esteem is built on evidence of specific achievement as a
mother, it would seem very likely to be influenced by "failure
as a mother" as seen by having a "difficult" child. Mothers
of such children no doubt would feel anxious and threatened
out of fear that the difficulties that they are having in
caring fer their infant are exposing their inadequacy as
parents. As f0r mothers of children at "biological risk",
they too, perhaps, are experiencing the "feeling of failure"
associated with not having delivered that "perfect child"
spoken of in the theory section of this study. It should be
noted that these factors - difficult temperament and high
risk status - were in and of themselves related to low self-

concept. The interaction of the two of them did not affect

76

self-concept any more than either of them alone. Thus, there
was no indication that having a "difficult child who was at
biological risk" increased the likelihood of the mother
having a lower self-concept, than if either of these factors
existed separately. It should also be noted that these two
variables accounted for twenty-three percent of the variation
in the self-concept of the mother. Thus, there are clearly
other factors involved here. Nonetheless, it does seem that
twenty-three percent is an influential amount and should be
recognized as important.

Until now, the relationships that have been noted in
this section have been examined from the point of view of
the child influencing the mother's self-concept. Since cause
and effect has in no way been determined, it would be essen-
tial to consider the relationship in the opposite direction.
This is to say that perhaps it is the case that the mother
has a lower self-concept prior to birth and, as such, influ-
ences the child's temperament. This might, hypothetically,
take place through attitudes and behavior that could adversely
affect the mother's ability to relate to her infant and cause
the development of certain styles of behavior in the child.
Mothers more concerned with having their own needs met, than
with meeting the needs of their child, would seem to be an
example of a situation where negative influences on the
behavioral style of the child could occur.

Examining the relationship between risk status in the
infant and the self-concept of the mother (in the direction

of effect noted above) would seem to be related to studies

77

carried out (see MacDonald, 1968) that have found evidence

to suggest that mothers who were more emotionally troubled,
more dependent, and had more negative attitudes towards preg-
nancy, had a greater incidence of premature births. Sameroff
and Zax (1973) also found that there were relationships be-
tween emotional illnes and prenatal complications, noting
that the length of time the woman had been ill rather than
the severity of the illness led to prenatal difficulties.

An interesting possibility related to the above issue is
the consideration that the self-concept of the mother in-
fluences the "biological risk status" of the child. Research
shows that how people feel about themselves is related to how
they feel about their bodies (Rosen et al., 1968). This
being the case, it may be that mothers who have negative
feelings about themselves as persons would also have negative
feelings about their bodies. Consequently, it would seem
feasible that mothers with negative feelings about their
bodies would take poorer care of themselves. This might take
the form of smoking excessively, high intake of alcohol and/
or drugs, or poor nutritional care. The fact that a mother
had a poor diet during pregnancy has been found to negatively
influence the prenatal environment provided for the infant
(see, for example, Birch, 1971), and to create a situation
where a child "at risk" could be born. An attempt at inves-
tigating the relationship between physical self-concept in
mothers and the "biological risk" status and temperament

pattern of the child was made in this study. Using the

78

physical self-concept scale of the Tennessee Self-Concept
Scale, self-concept was examined in relation to these two
variables. Significant differences were not found between
the physical self-concept of mothers in the "high risk" group
and those in the "average" group. However, there were signi-
ficant differences in self-concept of mothers of infants in
the different temperament clusters. Mothers of infants at
the "difficult" end of the temperament continuum were found
to have lower self-concepts than mothers of babies at the
"easy" end. This seems to be another indication of the rela-
tionship between self-concept in the mother and the tempera-
ment of the infant. The fact that there was not a signifi-
cant difference in the physical self-concept of mothers in
high risk and average populations would seem to speak against
the possibility of mothers' negative feelings about their
bodies influencing their child's biological risk status. It
would appear that this issue is a highly complex one and
findings noted here are far from conclusive. This point
ranains to be investigated in greater detail in the future,
possibly using additional research methods and instruments.

Another factor examined through the data in this study
was the Family Self-Concept variable of the Tennessee Self-
Concept Scale. This factor was purported to reflect the
mother's feelings of adequacy, worth, and value as a family
member. It was found to be significantly related to tempera-
ment and biological risk status of the infant, thus further

focusing the cause and/or effect of self-concept on these

79
two variables. This again was in the direction that was
expected (i.e., with mothers who had difficult infants having
lower self-concept scores). Statements made about the effect
of temperament and risk status on the Family Self-Concept
could be made with greater certainty than the same statements
made about the overall self-concept (as measured by the Total
Self-Concept score on the T.S.C.S.). It would appear that
there is less of a sense of worth or value as a family mem-
ber in the mother when either difficult temperament or high
risk status is an existing condition in her child. Consider-
ing what would be involved in caring for a difficult child,
it would seem that one's sense of adequacy and worth could
very likely be affected. Furthermore, when one realizes the
emphasis placed on the mother's role in our society and her
importance as a family member, it would appear that this par-
ticular variable would be quite significant and consequently
reflect the critical nature of the self—concept/temperament,
biological risk relationship.

Certainly, the possibilities thus far examined in this
chapter, representing opposite points in the "direction of
effect" issue, are far from exhaustive. Other variables are
likely involved in the relationship between the three varia-
bles discussed. It would appear to this writer that a plau-
sible response to this "direction of effect dilemma" is that
there is an interaction between a temperament predisposition
in the infant and the self-concept of the mother. Thus,

rather than the self-concept simply affecting the infant

80
(and his temperament), and the infant's temperament simply
influencing the mother's self-concept, there is a mutually
influencing interaction pattern affecting each dyad to differ-
ing degrees and in a multitude of ways. It is also possible
that there is an interaction between mothers' self-concept
and the biological risk status of the infant and that this
too is not simply a unidirectional reaction.

Attempting to get further insight into the relationship
between self-concept, temperament cluster, and risk status,
mothers in this study were asked about their infants' in-
fluence on their feelings about themselves. In regard to
this question, as was expected, the majority of the mothers
thought that their child had, in fact, influenced the way
they felt about themselves in some way. Of these mothers,
an even greater proportion thought that they had been in-
fluenced in a positive way. Of the mothers who thought that
they were negatively influenced, most had children rated as
difficult on the Infant Temperament Questionnaire. Carey
(1970) had noted that the more difficult the infant was, as
measured by his questionnaire, the more likely the mother was
to report having found his temperament a problem. Thus, the
fact that most of the mothers who felt negatively influenced
by their child had "difficult children" (as measured by the
Infant Temperament Questionnaire) would in a sense relate to
Carey's observation. Mothers who have found their child's
temperament to be a problem would, it seems, also be more
inclined to feel that their child had negatively influenced

the way they felt about themselves. This point was by no

81

means found in all the mothers of difficult infants. What
about the remainder of the mothers of difficult babies? Did
they actually perceive their infants as having positively
influenced the way they felt about themselves? It seems that
this could be interpreted in several ways. The first is that
the "positive" aspects of their child and their relationship
to him outweighed the negative points and the majority of
mothers of "difficult" babies did indeed feel positively
influenced. A second possibility was simply differences in
frustration tolerance in the mothers. Perhaps these mothers
of "difficult" infants develop a higher frustration tolerance
in response to their children, and thus learn to deal with
them, showing a minimal amount of stress. A final explana-
tion, that appears to be the most plausible to this writer,
is the social desirability factor. Quite possibly, mothers
in general were reluctant to admit that their child could
influence them in a negative fashion. This goes along with
McBride's (1974) observations of the metaphor of mother equal
to "perfect, love, and nurturance". Our society has conveyed
the idea that mothers are always loving and accepting under
all conditions and at all times. Admitting that their child
had influenced them in a negative way would very possibly be
equated with being a failure as a mother.

It is important to note that mothers who were "positively
influenced" by their child had a significantly higher self-
concept than mothers who were "negatively influenced". Thus,

'mothers who felt that they were adversely affected by their

82
infants were the same mothers who felt poorly about themselves
in general. Considered together, these factors strengthen
the argument for the effect of the child on the mother's
feelings about herself.

Another approach to gaining insight into the relationship
between the variables discussed was to have mothers and
psychologists rate the temperament of the infant and then to
compare the results. These results were also compared with
the more detailed and objective scores on the Infant Tempera-
ment Questionnaire. Mothers were found to be in agreement
with the Questionnaire scores 43% of the time. However, in
42% of the cases. the mother rated the child as easier than
was indicated on the Questionnaire. Here again we can see a
tendency to minimize the difficulty of the infant. Of the
twenty-three infants who were found to be "difficult" tem-
peramentally (as measured by the Infant Temperament Ques—
tionnaire), only six were acknowledged as such by their
mothers. The explanations just provided for this similar
trend in response to the "how influenced" question would
appear to be applicable in this situation as well. This does,
after all, appear to be a similar type of inquiry. It was
observed, however, that mothers were less reluctant to call
their child "difficult" than they were to say that they had
been "negatively influenced" by him. Looking at this finding
from a social desirability perspective, one might speculate
that this relates to the issue of being in control, which is

so highly emphasized in our society. For example, McBride

83
(1974) notes that mothers are expected to act in controlled
ways and to be able to shape the child into an "outstanding
product" and through their control alone create children who
conform to our society's stereotype of what is a "good" child.
If they are not in control, this would seem to have negative
connotations. Having a child that could create negative
influences in themselves would seem to be just such a situa-
tion and they may be less willing to admit this effect.

In comparing the mother's perceptions of the child's tem-
perament with the psychologist's perceptions, we find that
the mother was inclined to rate the child as easier than the
psychologist. Attempts at explaining this trend are more
complicated than the comparisons that have just been.made.
These points will be discussed later on in this chapter when
methodological issues are addressed. At this time, it would
suffice to note that psychologists are evidently more objec-
tive and have a wider range of infants with which to compare
each infant. Mothers may have assumed that their child was
"as easy as others", since this question did ask the mother
to compare her child to others his age. Psychologists were
also apparently less reluctant to call a child "difficult",
and also had stricter guidelines for what constitutes an
"easy" child. Both of these points relate back to the objec-
tivity issue. Psychologists were more accurate in rating the
infant's temperament (again, if accuracy is defined as agree-
ment with the Questionnaire), but there was a noticeable
trend to rate the child as "average". Apparently, when

mothers were in doubt about which temperament category a

84

child fit into, they classified him as "easy". When psycho-
logists were in doubt, they rated the infant as "average".

After examining the ratings of mothers, psychologists,
and scores on the Infant Temperament Questionnaire, the idea
of "accuracy of perception" would be a helpful one to consi-
der. Most notably, mothers of "difficult" infants were found
to agree in only f0ur percent of the cases with the scores on
the Infant Temperament Questionnaire, while in the "easy"
temperament clusters, the mothers agreed seventy-three percent
of the time. This extreme discrepancy might be interpreted
as evidence that mothers of "difficult" infants were more
likely to misperceive their infant's temperament than mothers
of "easy" babies. While this is strictly conjecture, it
could be that mothers who are reluctant to acknowledge their
child's temperament might also be reacting to them in ways
that fail to take into account the child's individual style
of behavior. This "reaction" may occur involving two poten-
tially harmful types of reasoning on the part of the mother:
1) "If my child is just as easy as any other child, and
things are going wrong, then it must be my fault."
2) "Since my child is just as easy as any other child, he
should do 'such and such' without any problem." These two
situations could very likely set the stage for future prob-
lems in both the infant and the mother, due to a failure to
recognize the child's individual temperament style. Not
recognizing a "difficult" infant as being difficult and
assuming that he is "just like any easy baby" is likely to

85
place considerable strain on a mother/child relationship.
Thus, a situation might be highlighted, whereby intervention
could dissipate a potentially harmful reaction. Intervention
could obviously occur only after the situation was identified.
The results of this study suggest that using an instrument
such as the Infant Temperament Questionnaire may be helpful
in identifying motheréinfant dyads where intervention may
prove constructive.

Some of the results of the limited amount of prior work
in the area of infant temperament research have been con-
firmed in this study. As in previous studies, there were
only modest differences between temperament scores of boys
and girls. While boys were found more frequently in the
"difficult" cluster and girls more often found in the "slow-
to-warm-up" and "easy" categories, these differences were
not statistically significant. Thus, there is no compelling
evidence to suggest strong sex-related genetic differences in
temperamental style. Although prior research has shown that
parents treat boy and girl infants differently (Korner, 1973),
this did not seem to influence temperament style appreciably
for either sex.

The relationship between birth order and temperament
cluster was not statistically significant, which would sug-
gest that ordinal position does not have a striking effect on
overall temperament style. Prior research would lead one to
expect that since the first pregnancy is believed to be a

"crisis situation" (Ryder, 1973), and that since "the second

86
child is so much easier than the first" (Thomas and Chess,
1977), then the first child should be more difficult to raise.
One would therefore expect that firstborns would have a
greater likelihood of being "difficult" in temperament as
well. This did not prove to be the case in this research
project.

Temperament clusters were found to be significantly
related to scores on the Denver Developmental Screening Test.
While the overwhelming majority of infants tested in both
populations in this study were well within normal limits,
about one fourth of the infants seen were either questionable
or delayed in development. Furthermore, the majority of this
group were from the "biological risk" population. Analysis
of the data indicated that infants with questionable or de-
layed development were overrepresented in those temperament
clusters towards the "difficult" end of the temperament con-
tinuum (particularly in the slow-to-warm-up category). One
possible explanation f0r this was the low activity level in
these infants, which is a component of the slow-to-warm-up
temperament cluster. Perhaps the motor problems in the de—
layed or questionable infants accounted for this temperament
classification. On the other hand, it is conceivable that
these infants were just "slow-to-warm-up" to the testing
situation and this accounted for their "delayed" scores. In
any case, there were a greater number of infants with develop-
mental problems in the more difficult temperament categories.

In this particular study, knowing that a child had a

8?
developmental problem made it a likely guess that he also had
a relatively "difficult" temperament. Facts such as these
lead one to question the possible effects on temperament of
developmental delays as the infant matures. A follow-up
study of these delayed infants with "difficult" temperaments
would be valuable in gaining greater insight into the poten-
tial importance of this relationship. This not only to
determine a possible cause and effect between the variables
but also to examine the possibility of confusing developmen-
tal delays and difficult temperament. It would appear that
it could be quite easy to mistake the inactivity, negative
mood, and slow-to-warm-up temperament of an infant with the
characteristics of a child who is developmentally delayed as
measured by the Denver Developmental Screening Test. It
stands to reason that confusion of factors associated with
"difficult" temperament and developmental delay could lead
to the implementation of less effective strategies of inter-
vention. The ramifications of such errors in examining the
child would likely have strong consequences for his future
healthy development, and consequently demands continued

future research.
Practical Issues

While there were several practical issues that recurred
throughout the life of this project, none was more pressing
and omnipresent than the dilemma of carrying out research
while maintaining a full concern for the ethicality of the

study. The complex administrative and institutional

88
arrangements understandably in effect to protect patients
were often frustrating. While stringent ethical views were
maintained throughout the project, it did appear that this
was often a hindrance. What would have ordinarily been excel-
lent opportunities for research were discouraged by strict
screening procedures. Thus, scientific concern had to be
disregarded in several instances in favor of social concern.
This clearly is a dilemma and it appears to be becoming more
complicated as increasing demands for ethicality are made.

Persistence on the part of those doing the research and
striving for scientific rigor in the face of numerous admin-
istrative regulations is the formidable goal for which the
researcher must aim. This, while at the same time he is
maintaining an ever increasing concern for the ethicality of
his project and the procedures being used.

Juxataposing this problem with the welcome reaction to
obtaining information on the part of all involved was quite
striking. Nowhere was this more evident than in the subjects
themselves. The majority of the mothers, and for that matter
the fathers, were anxious to learn more and more about their
new infants. This curiosity dominated interviews with par-
ents, who were more than too happy to talk, share, and learn
(and very often - teachl). It seemed that the mothers were
unusually conscientious in filling out the designated ques-
tionnaires and eager to obtain results. This idea of re-
ceiving results and the strong curiosity that was evident,

seemed to be the primary motivation for taking part in a

89
study that provided no financial reimbursement. These in-
centives likely helped account for the fact that no mother
refused to take part in the study and only twelve percent of
the mothers of "average" infants, and eighteen percent of the
"high risk" infants, failed to mail back their completed
questionnaires. This high return rate was important to note
since one of the most serious issues raised in using ques-
tionnaires is the bias that could result due to nonresponse
(Hillestad, 1972). Evidence has been found to indicate that
very often nonrespondents differ from those who do return
their completed questionnaires. Hillestad also notes that
unless the response rate is over eighty percent, sample bias
is likely to be quite prominent.

Other factors were also believed to have contributed to
this high rate of volunteer participation. Rosenthal and
Rosnow (1975), noting that situational determinants of volun-
teering are very important in regard to methodology, have
observed that in addition to an interest in the topic under
investigation, perceiving the investigation as important was
influential in encouraging volunteers. Most of the parents
recognized the practical significance of this study and the
possibility of it benefitting others. This point was
strengthened by their physician's endorsement of the project
(or the clinic physician's endorsement). The mid-Michigan
communities where the research was carried out were well
accustomed to requests for research participation, and while

this could be looked at as a hindrance (i.e., due to the

90
possibility that potential subjects would be driven away
because they felt "over-researched"), this was not the case.
The positive reputation of the University and the department
within the University where the research was carried out, no
doubt contributed further to the high rate of response, as
did the fact that volunteering was treated as the appropriate
thing to do. Most importantly, the efforts to make the inter—
actions with the parents as informal, nonthreatening, and as
individualized as possible were all believed to be important

positive factors.
Methodological Issues and Limitations of the Study

The first methodological issue warranting mention con-
cerns the rating of temperament, and specifically the use of
the Infant Temperament Questionnaire for this purpose. Find-
ings in this study would speak to the value of using this
instrument. Comparing the frequencies of the various tem-
perament clusters in the New York Longitudinal Study, Carey's
standardization sample for the questionnaire, and the current
study, similarities can be noted.

Simply asking mothers and/or psychologists to rate a
child's temperament in general provided agreement in about
one half the number of cases seen. Various complications
result if one uses just one of these reports. It should be
noted that only two psychologists were used in this study and
no claim is made that they are representative of the popula-
tion of psychologists at large. Nonetheless, there are cer-

tain factors that can be observed that likely generalize to

91
psychologists and other workers who might evaluate infant
temperament. One of these issues is the fact that the psy-
chologist is seeing the infants under far from ideal condi—
tions. Infants are frequently very fussy in doctors' offices
(and in clinics) and as such, to use one mother's words, the
babies "are not themselves". In addition, the psychologist
does not have extensive opportunity for observation of the
infant. In both the clinic and in this research study, the .
amount of time the psychologist was able to spend with the
infant was realistically limited and longer periods over
extended periods of time would be needed for the most exact
estimates of temperament.

As for the mothers, the question of accuracy in the use
of their general descriptions of their child's temperament
has already been noted. The lack of objectivity due to
heavily value ladened questions, and the generally limited
frame of reference they have for infant temperament are part
of the difficulty. Inaccurate recollections, which reflect
the mother's biases about the infant's behavior - rather than
the actual behavior - further complicate the problem with
using mothers' general temperament ratings.

The Infant Temperament Questionnaire does overcome some
of the problems with objectivity, and while the difficulty
with ambiguity of the frame of reference would still be an
issue, the problem is considerably less with this instrument
due to the fact that the behavior in the questions has been

broken up into relatively specific observable and measurable

92
traits. While other problems in rating temperament such as
central tendency errors (i.e., those cases where mothers
avoided using extremes and stuck to middle ratings) were
still likely to be issues when using this scale, it does pro-
vide the psychologist with greater insight into the child's
temperament.

Carey had received some initial criticism regarding his
original Questionnaire (1970) and very positively went about
correcting these faults. Increasing the number of items and
rating options, randomizing the items as to content area and
category, and clarifying questions by using simplified terms
have been answers to these comments and have strengthened the
instrument. In addition to these improvements in the Infant
Temperament Questionnaire, certain steps were taken in the
procedures in this study to insure a high degree of validity.
Efforts were made to motivate the mothers to do as accurate
a job as possible. If they felt that the topic under study
was about some inconsequential matter, it was believed that
they might have answered the questions by marking anything or
giving little thought to their responses. Because of this,
the mothers were given the questionnaires to complete at home.
This was done in part to allow them.more time to think about
their answers than would have been possible if they were
interviewed or if they simply completed the form while wait-
ing at the clinic. They were also encouraged to answer only
those questions applicable to their child and were given the

researcher's phone number and address so that they could

93
contact him if they had any specific questions.

The methodological issue of questionnaires versus
observational data has been addressed in many studies (see
Thomas and Chess, 1977; Scholom, 1975). Thomas and Chess
have stressed that the accuracy of behavioral data is related
to the emphasis on objective descriptive reports from a wide
variety of sources, life situations, and experiences. Clear-
ly, there is a need for the development of increasingly
sophisticated observational techniques for use in various
situations. Considering what has just been noted, it is
believed that the Infant Temperament Questionnaire, while
having many of those shortcomings attributed to question-
naires, has been proven to be quite reliable and valid for
the purpose of examining temperament. While being relatively
specific and objective, it is also economically feasible (in
both time and money) for use by physicians, nurses, psycholo-
gists, and mental health workers dealing with infants, not
to mention for use in furthering research studies such as
this current one.

Another methodological issue relates to the measurement
of self-concept. There are researchers who believe that self-
perception cannot be assessed. There does seem to be some
confusion and disagreement about what exactly "self-concept"
actually is. Wells (1976) has observed that dealing with
"psychological ideas" such as self-concept involves a number
of complex issues. As used, self-concept is a "hypothetical

construct formed by social scientists to summarize certain

94
features of human behavior and it shares all of the diffi-
culties that people have encountered with such scientific
abstractions" (p. 9). Nonetheless, the utility of the theory
of self-concept has been demonstrated in the countless num-
ber of research studies done over the years. This concept
has been found to be extremely useful for describing and
explaining human behavior.

Since self-concept is theory bound, each usage of the
term is relative to the particular theoretical context in
which it occurs. This is to say that it is from each theory
that self-concept gets its definition. Wells (1976) notes
that certain basic analytic properties must be present if one
is to be able to describe self-concept in a way that allows
comparisons among descriptions. In terms of an instrument,
this means that the observations "should be standardized,
objective, and quantitative" (i.e., numerically expressible
in terms of the amount of self-concept reflected) (Wells,
1976).

Keeping the above in mind and recognizing the difficul-
ties associated with measuring self-concept, it was decided
that rather than avoiding the topic, further research in the
area was needed. Recognizing that the Tennessee Self-Concept
Scale represents "one of the best of the current scales
specifically designed to measure self-concept" (Robinson and
Shaver, 1976), it was chosen. This is not to say that all
problems are eliminated by using this instrument. Clearly

some remain, the most obvious of which is the control of

95
social desirability. Despite the fact that a "lie scale" is
included in the Tennessee Scale, the problem of mothers
answering to appear socially appropriate was no doubt an
impinging factor. This, despite the fact that no names were
used throughout the study and mothers were emphatically en-
couraged to put down their "real feelings".

A final point to stress is that low self-concept scores
must be considered carefully. Different people derive esteem
from widely different sources. Thus, low self-concept in
mothers with "difficult temperament" or "high risk" infants
might come from other sources not studied in this project,
and it is likely that it comes from an interaction with such
variables. The family self—concept score on the Tennessee
Self-Concept Scale was an attempt at controlling for this
point and it is believed that it did add strength to infer-
ences made about the relationships.

It is hoped that this research study has generated new
questions and shed some new light on timeworn issues that
have been examined in the past. In any case, hopefully this
work will have heuristic value for the encouragement of addi-
tional research in the areas under study at this time. Some
of these possibilities for future exploration will be

examined in the following section.
Future Research

A major possibility for future research focuses on the

particular designs of the studies. This cross-sectional

96
study does not offer any proof of cause and effect in the
relationship between the variables investigated. In no way
is this study stating that the mothers' self-concepts are
lowered by difficult infants or that difficult infants are
created through the low self-concepts of the mothers. A
longitudinal study would be beneficial in this regard. If
mothers could complete the Tennessee Self-Concept Scale prior
to the birth of their child and these same mothers "followed
up" at several points after the child's birth, it might be
possible to observe any changes in self-concept in correla-
tion with the temperament of the infant. This would lend
strength to the cause and effect argument but by no means
eliminate the question of innate constitutional differences.
Blank (1976) notes that even within a few days after birth
the infant's behavior may be influenced by the environment:
thus, it is extremely difficult to interpret innate differ-
ences.

It would also seem to be very valuable to follow up
populations such as those under study at this time in an
effort to explore the development of behavior disorders in
children, specifically as this relates to temperament, bio-
logical risk status, and the self-concept of the mother.

An interesting idea that occurred to this writer while
carrying out this study related to the broader aspect of
parent-child interactions. It would appear to be quite
valuable to speak with parents prior to the birth of their
child and have them "paint a picture" of the baby as they

97
would like him or her to be. One year after birth it would
be interesting to approach these parents again and note their
ideal infant picture and whether or not their infant did fit
with this picture or whether they had changed their idealized
image of the child. Solnit and Stark (1961) have observed
that during the course of a normal pregnancy, the mother and
father develop a mental picture of their baby. They note
that one of the tasks of parenting is to resolve the dis-
crepancy between the idealized image of the infant and the
actual appearance and behavior of the real infant. One would
hypothesize that there would be a degree of modification on
"both sides". This is to say that parents would likely change
their image to correspond with the actual infant, and the in-
fant would likely change in the direction of the parents'
image. Considering the populations studied in this research
project, it would seem that a mother with a premature infant
would have "further to go" to adjust her image considering
the physical state of most premature infants (i.e., small
in size, poor health, etc.).

Another point that would seem worth studying, would be
the interactions between both of the parents of a "difficult"
temperament child. Is there an impact on the family as a
system? Does such an infant create tension and disagreement
over handling between his parents? Are the siblings affected?
If so, how?

One occurrence that was noted informally in this study

was the fact that the mothers of "high risk" infants reported

98
that their child had been very "difficult" (this in response
to questions about the child's temperament). However, as
the infant got "older", and at the time of this study, the
mothers observed that their infant had become "considerably
easier". This could possibly explain why more "difficult
infants" were not found in the biological risk group. Per-
haps any change in the direction of "easier" temperament was
perceived as placing the child in the "easy" category. These
parents conveyed the feeling, as one mother put it, that
through their "persistence and sense of humor" they were able
to "manage" their child. Thomas and Chess (1977) note that
the first eighteen months are the hardest. It would be an
important finding if trends were discovered that did, in
fact, confirm these speculations. Do infants become easier
in their temperament style or is there simply a mutual adjust-
ment on the part of mother and child?

Finally, future research should have as a goal the
development of more rigorous instruments for measuring tem-
perament. Since the value of temperament as a significant
factor in the development of children has been established,
more efficient devices for examining it would seem essential.

This was only an introduction to the possibilities for
future research. It seems to this writer that the possibili-

ties are endless and the need clearly demonstrated.

99
Implications

Several salient, practical implications can be inferred
from this study. These implications relate not only to the
temperamental individuality of infants and the self-concept
of mothers but to other areas, including the self-concept of
children, child abuse and intervention, and the idea of
parenting versus mothering.

In this study, as in others in the past, the tempera—
mental individuality of infants has been demonstrated. This
is, however, one part of a complex and organized human being
with abilities that are more developed than most people would
believe, even at a few months of age.

There are several areas which together make up the total
child. These include intellectual, physical, and sociaL/
emotional components. If a child is to achieve his full
potential, all areas must be considered and helped to develop.
Clearly, the results of this study lend strength to the be-
lief that along with considering physical health, cognitive
abilities, and social interactions, the temperament of the
child must also be examined. It is apparent that there is a
need on the part of parents, and later on the part of schools,
to respect the individuality of the child. This includes not
only his growth patterns and innate aptitudes, but also his
own specific temperament.

This issue would seem to relate to child rearing prac-

tices in general. It would appear that allowances must be

100
made for differences in temperament. Too often in the past,
research has been carried out assuming that children are
pretty much all alike, so that one kind of behavior on the
part of parents would affect each child the same way. 0b—
viously this is not so, considering the differences in tem-
perament observed in the infants in this study. It would
appear that child rearing practices must be considered along
with the temperament of the child. It would seem that there
is, in reality, no one best way to bring up a child, but
rather a style of parent child rearing geared to the parti-
cular infant and his temperament.

In this study, the temperament of infants was found to
be related to the self-concept of their mothers. It has been
noted that the basic attitudes that make up the child's self—
image develop out of his interactions with his immediate
emotional environment. This relationship between infant
temperament and maternal self-concept would likely influence
the growth of self-concept in the child. The child's self-
concept is first and foremost an outgrowth of his home and
the people who live with him there. ‘While the focus in this
study has been on the effect of temperament of the infant on
the mother's self-concept, certainly the implication goes
beyond the mother to the child's self-perception. This is
to say that the child's temperament and risk status could
directly and indirectly influence his mother's self-concept
and in turn his own. If a child's parents have not been able

to demonstrate an attitude of acceptance towards the child

101
and allowed him some flexibility in his growth, possibly
because of his "difficult temperament" style, then quite
likely the child could develop a negative self-concept.
Furthermore, if he is given the impression that he is not
worthwhile, because his temperament is more difficult than
his parents would like, this too could possibly lower his
self-concept. As has been noted in the theory section of
this study, Coopersmith (1967) has found evidence suggesting
that children with low self-concepts had mothers who were
themselves low in self-concept. He noted a significant
modeling effect which found children likely to identify with
their "active, poised, and relatively assured" high self-
concept parents. A child with a difficult temperament style
and with parents with low self-concepts would evidently have“
less of a chance to imitate such behavior. It would seem
that this topic of self-concept in the child would relate to
a study by Cox (1966) which found evidence to suggest that
parental disagreement was significantly related to negative
self-concept in the child. The implications for this present
study are that if a "difficult" child causes discord and
disagreement between his parents (see section on Future
Research), then the repercussions of this strife might be
evidenced in his self-concept. Thus, it appears that there
are very real possibilities of maternal self-concept and the
temperament of the infant influencing the child's own self-
concept.

There has been an emphasis in this study on the fact

102

that mothers are not born with special skills that automat-
ically enable them to handle all infants without difficulty.
"Motherhood is a role that one must grow into gradually"
(McBride, 1974). Complicating matters is the fact that our
society is preoccupied with the idea of "perfect mothers
producing perfect children" (McBride, 1974). McBride goes
on to suggest that we should be concerned with what is
"appropriate and better" rather than perfection. This point
would relate to the idea discussed earlier, of mothers being
"blamed" for everything that could go wrong with their in-
fants, including what has been recognized in this study as
"difficult temperament". The speculation in this research,
which was not disproved, was that this attitude could be
related to the mother's self-concept and quite possibly set
the stage for a cycle of impossible expectations on the
mother's part and guarantee failure. If a mother has expec-
tations of "making a difficult child conform" to stereotypic
images of a "good baby", then she is likely to meet with
failure. If she in turn assumes that this failure has
occurred because of her incompetence as a mother, it very
possibly could lower her self-concept. As has been noted
(see Kiesler et al., 1970), this could affect her actual
performance as a mother and serve only to increase the like-
lihood of failure.

Earlier it was stated that cycles like the above are
conducive to abusive situations in the home. Child abuse

has been established as a major problem in our society.

103
Research has shown that it is a very complex issue, with many
variables affecting the outcomes of various situations. It
would seem, however, that the temperament of the child is,
in fact, one variable that rightly ought to be considered as
important. Recognizing those mother—infant dyads where ex-
treme conflict exists would be paramount in keeping situations
from reaching the abuse stage. How is one to recognize those
situations? From this study, it was implied that extreme
discrepancy between scores on a relatively objective tempera-
ment questionnaire, when compared with the mother's general
ratings of a child's temperament and a psychologist's general
ratings of a child's temperament, could lend insight into
potentially difficult situations in the home. Gregg (1973)
has found evidence to suggest that child abusing mothers
were more likely to perceive their children as more difficult
than they actually were. It would seem that these three
measures could likely compliment each other by giving a more
comprehensive view of the child, his temperament, and its
possible effect on the family situation. They could help
gain insight into the mother's perceptions of her child's
temperament compared to more objective measures of how "diffi-
cult" the child appeared to be. For example, a situation
where a psychologist and an objective temperament question-
naire rated a child as difficult and the mother felt that
the child was easy, would probably be looked at (and likely
should be handled) differently from a case where the psychol-

ogist, mother, and the questionnaire were all in agreement

104

about the child's temperament. The fact that the mother,
for example, was in disagreement with both the psychologist
and the questionnaire could indicate that she might be mis-
perceiving the child's temperament style. Quite possibly she
was having difficulty coming to terms with her child's indi-
vidual style of behavior and might need help in understand-
ing and adjusting to it. On the other hand, a mother in
agreement with the psychologist and the questionnaire about
her child's temperament would seem more likely to be recog-
nizing her child's style of behavior but might need support
in dealing with it. These are indicative of two possible
situations that could arise. It should be noted that the
fact that there is disagreement between both the psychologist
and questionnaire as compared to the mother's rating clearly
does not mean that the mother is definitely failing to recog-
nize her child's temperament style. It would, however, seem
to highlight situations warranting closer examination to in-
sure the mental health of both mother and child. Child Find
centers could be optimal locations for screening using instru-
ments such as those just discussed. Programs such as Child
Find could also follow up infants at planned intervals,
monitoring any changes in the status of the infants and their
relationships with their parents.

A critical issue that is related to much, if not all,
of the ideas discussed in this work is that of the concept
of mothering. Implications from this study would suggest

that in many instances, especially in the case of difficult

105
and high risk babies, it might be beneficial for both mother
and infant if the role of primary caretaker was shared.
While the father would be the most likely person for this
sharing, it could also involve other relatives if necessary,
or possibly some type of day care facility. Butler (1970),
among others, has noted that adjustment leading to a produc-
tive life is most facilitated if the child is cared for by
many warm, friendly people, so long as there is continuity
with a mothering person. It would seem likely that once
others are involved in the child's rearing process and the
responsibility for the socialization of the child is not
resting solely on the shoulders of the mother, feelings of
guilt and inadequacy would be less likely to develop in the
mother. In the past, the mother has often been held entirely
responsible for the child's problems and as such, this pro-
bably created many problems for both mother and child - not
to mention the entire family as well. No doubt much parental
discord has been generated by such situations. As Wortis
(1970) has observed, "recognizing the social needs of mothers
and infants, some investigators have begun to encourage entry
into day care nurseries at much earlier ages than is cus-
tomary". This is not to say that the family is not impor-
tant. Research and common sense have taught us that the
family is extremely important. Indeed, the family is so
important that opportunities should exist, particularly for
mothers of "biological risk" infants and infants with "diffi-

cult" temperaments, to have access to adequate child care

106
facilities. Mothers need their separateness from their child
and time for themselves. Time to collect their thoughts and
time f0r a much needed rest! Mothers need places to leave
their children when "crisis" situations arise and pressures
have built up to such an extent that they must get away.
Organizations such as Child Find could be starting points
where mothers could learn about the existence of facilities,
such as full and part time child care centers, where they
could stop in at "crisis" points, leave their children, and/
or talk about their own special needs. At these times, fur—
ther assistance could be offered in the form of complete
listings of what is available in the community in terms of
supportive services. The provision of nurses, social workers,
other mental health workers, as well as other parents for
home visits to help parents deal with difficult situations
is another option. A possibility that seems particularly
promising for helping parents to deal more adequately with
their child relates to "parenting programs". Parenting
refers to those skills needed by the adults who assume res-
ponsibility for the care and guidance of children. Presented
through classroom and group programs, Parenting Programs are
designed to disseminate materials, increase awareness, and
better educate caregivers in child growth and development.
The options for training could take place at various levels
including: high school family life courses, vocational
education courses for students who wish to prepare themselves

for careers as child caregivers, infant/parent programs for

107
young parents wishing to continue their education, continuing
education courses for older parents seeking to further their
parenting skills, as well as further training for profession-
als who work with infants and parents. In addition to pro-
viding knowledge about the way children grow and develop,
these programs could also provide information about the pro-
cedures necessary for parents to obtain services for them-
selves and their children when they are needed. These skills
would clearly be important in the case of "biological risk"
infants and with infants having "difficult" temperaments.
Unquestionably the parents of these infants need specific
knowledge, awareness, and understanding in order to help
their children grow up to be as healthy as possible.

In the sample of families studied in this project, there
appeared to be strong appreciation for the shared responsibil-
ity and pleasure of caring for children. Perhaps this sharing
of responsibility accounted for, in part, those instances
when there wasn't a significant lowering of self-concept in
the mothers of difficult and high risk infants. It was the
writer's observation that many fathers in these instances
helped their wives maintain positive pictures of themselves,
by assisting in setting realistic goals and developing favor-
able estimations of their performances as mothers. When
mothers and fathers equally share the responsibility, it
would seem that there would be less of a belief that the
course of development in the infant is solely determined by

the mother's attitudes, motivations, and needs. Consequently,

108

there would be less of a threat to the mother when the tur-
moil of raising a "difficult" child arises.

Some writers have recently gone so far as to suggest
that the term "mothering" should be "buried" (McBride, 1974).
The A.P.A. (1977) has suggested in dealing with problems of
designation, that the term "parenting" be used to describe
this nurturing that parents give. Moving from "mothering"
to "parenting" would, it seems, allow each of the parents
more freedom to appreciate their children more fully. As
Thomas and Chess (1976) have observed, "Parents who enjoy
their children in their own individual ways can be sure the
children are also benefiting. If there is one thing above
all that children will thrive on it is their parents' pleas-
ure in them" (p. 203). Continued encouragement for men to
take increasing parts in emotional and social interaction
with their infants and children has been found with growing
frequency in child development literature. This study would
directly and indirectly support this stand. As Wortis (1970)
has noted, this requires continued emphasis in social and
educational institutions to help prepare future generations
fOr this sharing of responsibilities. It would seem that

everyone stands to gain from trends in this direction.
Summary and Conclusions

In this research, no significant relationships were
found between temperament and 1) a child's "biological risk"

status, 2) a child's sex, and 3) a child's ordinal position

109
in his family. Thus, it would seem that none of these varia-
bles influence temperament in any striking fashion. One can
infer that any behavior or attitudes on the parents' part
that are specifically related to either of these three vari—
ables do not appreciably influence the infant's temperament.
This does not eliminate the possibility that over time these
factors could modify the child's temperament to a significant
degree. This point would need to be examined in a follow-up
study where changes could be studied. In any case, of those
early postnatal parental influences that do affect the child's
temperament, the above three variables are not among them.

On the other hand, "developmental level", as measured
by the Denver Developmental Screening Test, was found to be
significantly related to the temperament of the infant.
Infants who were delayed developmentally were more frequently
found in the "difficult temperament categories". It could
be possible that infants who are "delayed" do develop more
difficult temperaments. While this has not been proven
directly, it does seem to be an important point for centers
such as developmental assessment clinics to be aware of when
screening infants who are at risk for the development of
growth or learning problems. On the other hand, these find-
ings do relate to another possibility. Perhaps it is the
case that the less active, bland mood, slow-to-warm-up in-
fant is more easily confused with a developmentally delayed
child. Furthermore, a child with a more difficult tempera-
ment style might not really show what he can actually do

110
during the testing situation. These points should serve as
a word of caution for examiners at developmental assessment
clinics to be aware of, and would speak to the need for
expecially sensitive evaluations when both developmental
delay and difficult temperament are observed in an infant.
Developmental deviations are very difficult to detect in
young children, and while such problems are often overlooked,
one must also be careful that they are notheing confused with
a child's difficult temperament style. Further study of this
relationship is most definitely warranted, considering the
possible ramifications of decisions made for providing sup-
portive services to the child for his future development.

A major purpose of this work was to examine the tempera-
ment of the infant and the mother's self-concept to determine
whether a relationship exists. There was a significant dif-
ference in the self-concept of mothers in the different tem-
perament categories, with mothers of infants with more diffi-
cult temperaments having lower self—concepts. Although the
direction of cause and effect between the variables was not
established, it does appear to this investigator that this
relationship would most profitably be considered as a conse-
quence of a reciprocal interaction between.mother and child.
That is, the self-concept of the mother influences the in-
fant's temperament and, in turn, the temperament of the in-
fant affects the self-concept of the mother. As such, inter-
vention at several points could have beneficial effects

should problems be detected. Not only could assistance be

111
directed to bolstering the self-concept of the mother, but
help could also be offered to assist her in adjusting to her
infant's temperament. This could help her feel more success-
ful and consequently serve as a catalyst in developing a more
positive self—concept. By adjusting to the infant's tempera—
ment, the mother would likely be placing less stress on him,
and thus influence the way he is behaving to a positive de-
gree. One can see how a positive cycle could be established
through this cycle of events.

Many infants are now surviving birth who would have died
in earlier years. The development of sophisticated techniques
and instruments has helped to save the lives of many babies
and at the same time created an ever growing population of
infants at "biological risk". These infants warrant special
consideration not only because of their physical difficul-
ties, but also because of the possible social/emotional in-
fluence these complications could have on the babies them-
selves, and those in their environment. With this in mind,
another intention of this study was to look at the relation
between "biological risk" status and the self-concept of the
mother. A significant relationship was noted, with mothers
of infants at "biological risk" having lower self-concepts.
The direction of effect, however, was not established.

Should this lower self-concept exist as a result of the birth
of a "high risk" infant - a fact that remains to be proven
in future research - it would seem that hospitals should make

arrangements to deal with this as early as possible. The

112

factors commonly associated with the birth of an infant at
"biological risk", including the separation of mother from
infant, have been consistently recognized as potential prob—
lem points for the future development of relationships be-
tween the mother and her child. Helping mothers to come to
terms with the problems of their children would certainly be
important for the future healthy adjustment of both mother
and child.

Helping mothers to recognize and respect their child's
individual temperament style would seem to be a need that
could be implied from this study. Mothers of difficult in-
fants appeared especially likely to misperceive their child's
temperament, or perhaps were simply reluctant to acknowledge
it. The stereotypes conveyed in our society, both overtly
and covertly, about mothers and infants need to be dealt
with. Clearly, all babies are not "Gerber babies", mothers
are not always in complete control of their infants, and they
do not have "exclusive" power to meet their baby's each and
every need. Mothers need to be helped to recognize that they
are not solely responsible for their child's behavior, and
fathers need to be helped to accept increasing amounts of
responsibility for the care and raising of their child.
McBride (1974) notes that once an equal relationship is
arrived at between mothers and fathers in dealing with their
child, they are then "free to think about how to divide the
task of parenting in new ways that might develop the growth

potential of all concerned".

113

If one thing has been realized at the completion of this
research study, it is that infants are truly complex indivi-
duals differing in many ways. It is apparent that they are
very influential in their relationships with their parents,
and affect their parents as much, if not more than their
parents do them. Appreciation of these facts could make
childrearing considerably easier.

Finally, parents - and as a result of this study - speci—
fically mothers, need to deal with preoccupations of perfec-
tion. Images our culture has created of "perfect" mothers
creating "perfect" infants need to be challenged. To
quote Briggs (1975):

"All any of us can do is our best each day, but

that 'best' will never be perfect. And children

do quite nicely with fully human parents who

fall short here and there" (p. 306).

APPENDICES

APPENDIX A

Nine Categories of Temperament and Their Definitions

* Taken from: Thomas, Alexander and Chess, Stella, Tempera-
and Development. (p. 21-22)

1) ACTIVITY LEVEL - the motor component present in a given
child's functioning and the diurnal proportion of active and
inactive periods. Protocol data on motility during bathing,
eating, playing, dressing and handling, as well as informa-
tion concerning the sleep-wake cycle, reaching, crawling, and
walking, are used in scoring this category.

2) RHYTHMICITY - the predictability and/or unpredictability
in time of any function. It can be analyzed in relation to
the sleep-wake cycle, hunger, feeding pattern and elimina-
tion schedule.

3) APPROACH or WITHDRAWAL - the nature of the initial res-
ponse to a new stimulus, be it a new food, new toy or new
person. Approach responses are positive, whether displayed
by mood expression (smiling, verbalizations, etc.) or motor
activity (swallowing a new food, reaching for a new toy, ac-
tive play, etc.). Withdrawal reactions are negative, whether
displayed by mood e ression (crying, fussing, grimacing,
verbalizations, etc?) or motor activity (moving away, spit—
ting new food out, pushing new toy away, etc.).

4) ADAPTABILITY - responses to new or altered situations.

One is not concerned with the nature of the initial responses,
but with the ease with which they are modified in desired
directions.

5) THRESHOLD OF RESPONSIVENESS - the intensity level of
stimulation that is necessary to evoke a discernable response,
irrespective of the specific form that the response may take,
or the sensory modality affected. The behaviors utilized are
those concerning reactions to sensory stimuli, environmental
objects, and social contacts.

6) INTENSITY OF REACTION - the energy level of response,
irrespective 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.

114

115

8) DISTRACTIBILITY - the effectiveness of extraneous envi-
ronmental stimuli in interfering with or in altering the
direction of the ongoing behavior.

9) ATTENTION SPAN and PERSISTENCE - two categories which are
related. Attention span concerns the length of time a parti-
cular 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.

APPENDIX B

Data Sheet for Recording Information from
Mothers' Reports and PhysicianZClinic Records

Code Number

 

Sex M F

 

Firstborn Yes No

Chronological Age

 

Developmental Screening Index Normal Questionable Delay

Additional:

Do you feel that your child has influenced the way you feel
about yourself? Yes No

If so, has this been in a positive or negative way?

116

MICHIGAN STATE UNIVERSITY APPENDIX C

 

COLLEGE OF EDUCATION ° DEPARTMENT OF COUNSELING. EAST LANSING ° MICHIGAN ' 48824

PERSONNEL SERVICES AND EDUCATIONAL PSYCHOLOGY Gary Saltzman

Consent to Participate Forms for Mothers

Consent to Participate
Study of the Relationship Between

Infant Temperament Variables and
the Self-Concept of the Mother

I agree to participate, along
with my child, in the above mentioned study. The nature of
the study and the processes involved have been adequately
explained to me, and even though I give my consent at this
time, I understand that I may, at any time, withdraw without
penalty from the study and retract my permission to use any
information obtained.

I have been assured that all infbrmation will be held in
the strictest confidence, and have been offered an explana-

tion of the findings at the conclusion of the project.

 

Mother's Signature Date

117

APPENDIX D

Mother's general Impressions of Infant's Temperament

(Carey/McDevitt)

How would you describe your baby's temperament in your
own words? ‘

In comparison with what you know of other babies of the
same age, how would you rate your baby as to the follow-
ing criteria? (Circle one)

I.

II.

III.

IV.

VII.

Activity level - the amount of physical activity
during sleep, feeding, play, dressing, etc.

(1) high (2) medium (3) low

Regularity - of bodily functioning in sleep, hunger,
bowel movements, etc.

(1) fairly regular (2) variable (3) fairly irregular

Adaptability to change in routine - the ease or diffi-
culty with which initial response can be modified in
socially desirable way.

(1) generally (2) variable (3) generally slow
adaptable at adaptation

Response to new situations — initial reaction to new
stimuli, to food, people, places, toys, or procedures:

(1) approach (2) variable (3) withdrawal

Level of sensory threshold - the amount of external
stimulation, such as sounds or changes in food or
people, necessary to produce a response in the baby.

(1) high threshold (much stimulation needed)
(2) medium
(3) low threshold (little stimulation)

Intensity of response - the energy content of responses

regardless of their quality.

(1) generally (2) variable (3) generally mild
intense

Positive or negative mood - amount of pleasant or

unpleasant behavior throughout day.

(1) generally (2) variable (3) generally
positive negative

118

119

VIII. Distractibility - the effectiveness of external stimuli

(sounds, toys, people, etc.) in interfering with ongo—
ing behavior.

(1) easily (2) variable (3) non-distractible
distractible

IX. Persistence and attention span - duration of maintaining
specific activities with or without external obstacles.

(1) persistent (2) variable (3) non-persistent
C. How has the baby's temperament been a problem for you?

D. In general, temperament of baby is:

(a) about average
(b) more difficult than average
(c) easier than average

APPENDIX E

Infant Temperament Rating

 

PsyChologIst's Form

ACTIVITY LEVEL - How much does
child move around during the
testing?

APPROACH/WITHDRAWAL - How does
infant behave when presented with
something (or some situation) for
the very first time?

ADAPTABILITY - How easy or diffi—
cult is it to modify the child's
initial response?

THRESHOLD OF RESPONSIVENESS -
How much external stimulation is
necessary to produce a response
in the infant?

QUALITY OF MOOD - How pleasant
or unpleasant is the infant
during the testing?

DISTRACTIBILITY - How effective
is external stimuli in inter-
fering with the ongoing behavior
of the infant?

INTENSITY 0F RESPONSE - How much
"energy" does the child use in
responding?

PERSISTENCE AND ATTENTION SPAN -
Does the infant stick to what he
is doing with or without external
obstacles?

HIGH

APPROACH

ADAPTIVE

HIGH

POSITIVE

DISTRACT.

INTENSE

PERSIS.

MEDIUM

VARIABLE

VARIABLE

MEDIUM

VARIABLE

VARIABLE

VARIABLE

VARIABLE

In general, the temperament of the baby could best be

described as:
(a) about average

(b) more difficult than average

(c) easier than average

OBSERVATIONS:

120

LOW

WITHDRAWAL

NONADAPT.

LOW

NEGATIVE

NONDIST.

MILD

NONPERST.

APPENDIX F

Infant Temperament Questionnaipe
(for 4 to 8 month old infants)
revised, 1977
by William B. Carey, M.D., and Sean C. McDevitt, Ph.D.

 

 

 

 

 

 

Child's Name: Sex
Date of Birth: Present Age
Rater's Name: Relationship

to child

Date of RatingI

 

The purpose of this questionnaire is to determine the general

pattern of your infant's reactions to his/her environment.

The questionnaire consists of several pages of statements
about your infant. Please circle the number indicating the
frequency with which you think the statement is true for your
infant. Although some of the statements seem to be similar,
they are not the same and should be rated independently. If
any item cannot be answered or does not apply to your infant,
just draw a line through it. If your infant has changed with
respect to any of the areas covered, use the response that
best describes the recently established pattern. There are
no good and bad or right and wrong answers, only descriptions
of what your infant does. When you have completed the ques-
tionnaire, which will take about twenty-five to thrity min-

utes, you may make any additional comments at the end.

121

IJZZ

USIJG THE FOLLOWING SCALE, PLEASE CIRCLE THE NUMBER THAT INDICATES
OFTEN TEE INFANT'S REGRET AND CURRENT BEHAVIOR HAS BEEN LIKE THAT
DESCRIBED BI EACH ITEM.
Variable Variable
Almost usually usually Almost
defer Rafely doesSnct dogs Engquently always
' C
l. The infant eats about the sane alncst , 2 3 4 S , alnost
. amount of solid food (within 1 never ‘ ° always
oz) from day to day.
2. The infant is fussy On waking up alnost . 2 3 5 , alnost
and going to sleep (frowns, cries) never ‘ - ° always
3. The infant plays‘with a toy for . -
under a minute and then looks 315°?‘ 1 2 3 5 6 since.
for another toy or activity. -ever always
4. The infant sits still while watch- almost 1 2 3 5 5 almost
ing TV or other hearty acrivityu never always
3. The infant accepts right away any
change in place or position of alncst l 2 3 5 6 :i?°3t
feeding or person giving it. never ’37s
6. The infant accepts nail cutting ' almost 1 2 3 5 6 almost
without protest. never always
7. The infant's hunger cry can be , m
stopped for over a sin te by a_nost l 2 3 S 6 ‘lfOSt
picking up, pacifier, putting never alleys
on bib, etc.
8. The infant plays continuously fcr _ .
more than 10 min. at a tine with almost 1 2 3 5 6 ‘l‘°s“
a favorite toy. never always
9. The infant accepts his/her bath 1 ., -
any time of the day without 34°“ 1 2 3 5 6 ‘3‘“
resisting it. never always
10. The infant takes feedings quietly . . .
with nild expression of likes almost 1 2 3 S 6 ‘13“3t
and dislikes. never always
ll. The infant indicates disccnfcrt - -
(fusses or squirns)when diaper is aiggst l 2 3 5 6 :3328;
soiled with bowel novenent. n ‘ y
I . The infant lies quietly in the alnost . 2 3 5 6 almost
bath. never ‘ always
13. The infant wants and takes milk - - _
feeding: at about the sane tines fl:::? 1 2 3 7 6 :Iizsé
(Within one hourIfrcm day to day. “° ‘ 7
14. The infant is shy (turns away or , . , i
i - . - .4, _ a_ncs. - a_ncs.
clings . -o.he-I on nee.._g an never 1 2 3 p 6 always

other child f r the first tine.

EON

IJZB

2.
Variable Variable
Almost usually usually
never Ragely doe; not does
1 4

lS.The infant continues to fuss during
diaper change in spite of efforts
to distract -im/her with game, toy
or singing, etc.

16.The infant amuses self for ‘ hour
or more in crib or playpen (looking
at mobile, playing with toy).

17.The infant moves about much (kicks,
grabs, squirns) during diapering
and dressing.

18.The infant vigorously resists addi-
tional food or milk when full (spits
out, clamps mouth closed, bats at
spoon, etc.)

l9.The infant resists changes in
feeding schedule (l hour or more)
even after two tries.

20.The infant's bowel movements come
at different times from day to day
(over one hour difference).

Zl.The infant stops play and watches
when someone walks by.

22.The infant ignores voices or other
ordinary sounds when playing with a
favorite toy.

23.The infant makes happy sounds (ccos,
smiles, laughs) when being diapered
or dressed.

24.The infant accepts new foods right
away, swallowing them prcnptl .

25.The infant watches other children
playing for under a minute and
then looks elsewhere.

26.The infant reacts mildly (just
blinks or at rtles brieflyIto bright
light such as flash bulb or lett v
sunlight in by p“‘ing up shade.

27.i e infant is pleasant (smiles,laughs

when first arriving in unfamiliar
places (friend's house, store).

25.The infant gets sleepy at about the

almost
never

almost
never

almost
never

almost
never

almost
never

almost
never

almost
never

almost
never

almost~

never

almost
never

almost
never

almost
never

)almost

neve 1‘

almost

same time each evening (within & hr.) never

29.The infant accepts regular procedures

(hair brushing, face washin , etc.

at any time without protest.

almost
never

.J

.J

'J

l—J

’4

.4

.J

.4

UI

UI

U!

V!

Di

V]

VI

.3;

Almost

U3

0\

O\

O\

0\

0\

I“

always

almost
always

almost
always

almost
always

almost
always

abnost
always

almost
always

almost
always
almost
always
almost
always

almost
always

almost
always

almost
always

almOSt
always

almost
always

almost
always

12fl+

3.

. Variable Variable
Almost usually usually
never Rarely does not does F

l 2 3 4

30.The infant sits still (little 1 03‘
squirming)while traveling in car AESer' 1
seat or stroller. ‘

31.The infant's initi reaction to a ”vast
new baby sitter is rejection nevgr l
(crying,clinging to mother,etc.)

32.The infant keeps at it for many almost 1
minutes when working on a new skill never ‘
(rolling over,picking up object,etc.)

33.The infant moves much (squirms, almost
bounces, kicxs)while lying awake never 1
in crib. ‘

34.The infant objects to being bathed .
in a different place or by a diff- 2:32;“ I
erent person even after 2 or 3
tries.

33.The amount of milk the infant 1 cs‘ .
takes at feeding is quite unpre- never. "
dictable (over 2 oz.difference) “ ‘
from feeding to feeding.

36.For the first few minutes in a new almost ,
place or situation (new store or never ‘
home) the infant is fretful.

37.The infant netices(lccks carefully , cs‘
at)changes in the appearance or 3336:" 1
dress (hairdo, unfamiii r clOthing)
of the mother.

38.Th infant reacts strongly to foods, . 0st
whether positively (smacks lips,, at: fl“ 1

aughs, squeals) or negatively 3° e-
(cries).

39.The infant is pleasant (coos,smiles . i

. ,‘g g ., , . ,. since. 1
etc.Idu.-ng p-0ceccres l.ke na-- never
brushing or face washing.

40.The infant continues to cry in ,

4- 1 4.. a. r most
spite or severe. m.nu.es 0. never 1
scathing. '
4l.The infant keeps trying to get a alm
' ' . i a 4 . a 93‘ I
desired toy, wh.cn .3 out 0. never -
reach,for 2 min. or more. ‘

42.The infant greets a new toy with a almost ,
loud voice and much expression of never ‘
feeling(whether positive or negative)

43.The infant plays actively with , .
parents-much movement of arms, almos. 1

legs, body. never

IO

UI

Ul

U!

U]

U.

U!

‘0.

U1

UI

Almost
algays

O\

0\

0\

almost
always

almost
always

almost
always

almost
always

almost
always

almost
always

almost
always

almost
always

almost
always

almost
always

almost-
always

almost
always

almost
always

almost
always

1215

4.

Variable _Yariable .
almost usually usually Almost
hever Rarely does not does frequently always

1 ' 2 . 3 4 6
:4.The infant watches another toy almost ' almost
{3 “ ' . l 2 3 4 5 6 1
when o-iered even .nougn already never always
holding one.
45.2he infant's initial reaction at
+ _ e
home to approach by strangers is 3:323“ l 2 p 4 S 5 2:323;
acceptance. ‘ 7
16.2he infant wants do'time hats at
4:: -4- as... ( “1:. a, h ' d“"’- almost 1 2 A S 6 “5‘05:
---e-._g t_.es ove- - our --- never 3 - alwa's
erence) from day to day. I
17.The infant continues eating solid 1 1 ‘
foods without reacting to differ- nevgit l 2 3 4 S 6 alwzfs
ences in taste or consistency. J
tS.The infant cries when left to almost 1 2 3 4 5 6 almost
play alone. never always
0 ‘M- ¢ .9 -- .1 he 4

4,.:-e in-ant adjusts d-tn-n lO‘m.n. 1 ost 6 almost
to new surroundings (home, store, :gie, l 2 3 4 5 alwav
play area). ‘ 4

50.2he infant's daytime naps are about 1 1 i

. a_m st almos
the same lengtn from day to day nevgr l 2 3 4 S 6 always
(under one half hour difference). “

l.The infant moves about much during almost 6 almost
I 1 J ' l 2 3 4 S
.eedings (squirms, nicks, grass). never always
52.2he infant reacts (stares or startles)
to sudden changes in lighting (flash almost 1 2 3 4 S 5 almost
bulbs, turning on light . never ‘ always
53.2he infant can be soothed by talking almost almost
or games when sleepy. never l 2 3 4 S 6 always
= “ “J '4 .. ' ’ ‘*‘

,i.-he i_.ant o-splays muon -eei.n8 games: 1 q _ 5 glass:
(vigorous laugn or cry) during never _ e 9 4 5 always
diapering or dressing. ‘

“3.2he infant lies still when asleep almost 1 2 3 4 S 6 almost
and wakes up in the same place. never ‘ always

56.!h infant adjusts easily and a, « i

. . alnoSt . - almos
sleeps well within 1 or 2 days with never l 2 3 4 a 6 always
changes of time or place. ‘

1:" ”Me ‘w’m *- v-0 1' " ' ~n

-:.--e ll._n. .-ac.s to cnang-s in elm i 1- i

’ ’ 3......030
temperature or type of milk or sub- nevi?“ l 2 3 4 5 o always
stitution of juice. -

58.3he infant watches television for almost 1 2 3 4 5 5 almost
more than 5 minutes at a time. never ‘ always

59.2he infant can be calmed for a few _ _
min tes by being picked up, almost l 2 2 4 a 6 almost
played with, T.V., if fussing never always
about soiled diaper.

60.!he infant wants and takes solid almost 31:33:
3 3 ' d- ' d. a. A w a I
-ood .eedings a. acou. .he.sane never ‘ c 3 4 7 o always

time (within 1 hour) from day to day.

126

5.
6 Variable Variable
Almost 11 11
neze. haéeiy doe; not does 3
' 4

61.?he infant is content (smiles, coos)
during interruptions of milk or
solid feeding.

62.The infant accepts within a few
minutes a change in place of bath
or person giving it

63.The infant cries for less than one
minute when given an injection.

64. Th 3 infant shows much bodily move-
ment (kicks, waves ar ) when
crying.

65.2he infant continues to react to a
loud noise (hammering, barking dog,

etc.) heard several times in the

same day.
66.” he infant's initial reaction is
withdrawal t'rns head, spits out)

when consistency, flavor or temp-
erature of solid foods is changed.

67.1‘he infant's time of waiting in the
morning varies greatly (by 1 how
or more) from day to day.

65.The infant continues to reject dis—
liked food or medicine in Spite of
p T'cents' efforts to distract with
games or tricks.

69.The
touch

infant reacts even to a gentle
(startle, wriggle, laugh, cry).

70.2he infant reacts st Wong to

strangers: laughing or cryin

7l.2he infL actively grasps or
touches objects within his/her
reach (hair, spoon, lasses, etc.).

72.The infant will take any food
offered without seeming to notice
the difference.

73. The infant's period of g-eatest
physical activity comes at same
time of day.

74.-he in. ant appears bothered (c-les,
souirms) when first put down in a

different sleeping olace.
7S.-he infant reacts mildly to meeting
fami‘iar people (ouiet em- es or

no response)

almost
never

almost
never

almost
never

almost
never

almost
never

almost
never

almost
never

almost
never

almost
never

‘ almost

21679 2‘

almost
never

almost
never

almost
never

almost
never

almost
never

H
to

W“)

I.)

.4

.J

..J

fl

'4

'4

'4

.J

.J

.J

'4

uently

\II

VI

\Dl

UI

V!

U!

VI

{3

Almost

always
6
5 6
5 6
5 6
5 6
S 6
5 6
S 6
5 6
5 6
5 6
5 6
S 6
5 6'
S 6
5 6

almost
always

almost
always

almost

always

almost

'always

almost
always

almost
always

almost
always

almost
always

almost
always

almost
always

a inas-
always

almost
always

almost
always

almoSt
always

almost
alw ys

12??

be
, Variable Variable
Almost usually usually
never .Raiely doe; not doés.
l 4

76.2he infant is fussy or moody through- almost

out a cold or an intestinal virus.

77.2he infant wants an extra feeding at
a different time each day (over
one hour difference).

78.The infant is still wary or fright-
ened of strangers after l5 minutes.

79.The infant lies still and moves
little while playing with toys.

BO.The infant can be i tracted from
fussing or squirming during a pro-
cedure (nail cutting, hair brushing,
etc.) by a game, singing, TV, etc.

81.!he infant remains pleasant or calm

with minor injuries (bumps, pinches).

82.2he infant's initial reaction to
seeing doctor is acceptance (smiles,
coos).

85.The infant reacts to a disliked
food even if it is mixed with
preferred one.

64.2he infant plays quietly and calmly
with toys (little vocalization or
other noise).

85.The infantis fussy period occurs

at about the same time of day (morning,

afternoon or evening.)

66.The infant lies still during pro-
cedures like hair brushing or nail
cutting.

87.The infant stops sucking and looks
when he/she hears an unusual noise
telephone, door bell when drinking
milk.

68. The infant
with parent:

69.The infant is calm in the bath.
or dislike is mildly expressed
(smiles or frowns).

90.2he infant
a new food

pays attention to game

Like

requires introduction of
on 3 or more occasions

for only a minute or so.

Frequently
5
l 2

never

almost . 2
never ‘
almost .

l 2

never

almost 1 2
never ‘
almost

never 1 2
almost 1 2
never

almost . 2
never ‘
almost 1 2
never ‘
almost l 2
never

almost . 2
never ‘
almost 1 2
never

almost 1 2
never
almost 1 2
never
almost 1 2
never

1
almost 1 2
never

before he/she will accept (swallow)it.-

‘Jl

\Jl

\Jl

DI

VI

K”

V!

.r.

Almost

always
6
5 6
5 6
5 6
5 6
5 6
5 6
5 6
5 6
5 6
5 6
5 6
5 6
5 6

.5 6
5 6

almost
always

almost
always
almost
always
almost
always

almost
always

almost
always

almost
always
almost
always

almost
always

almost
always

almos:
always

almost
always

almost
always

almost
always

almost
always

1253

7.'

Variable Variable

Almost usually
never Rarely does not does
1 2 3 4

9l.The infant's first reaction to any new
procedure (first haircut, new medicine,
etc.) is objection.

92.!he infant acts the same when the
diaper is wet as when it is dry.
(no reaction)

93.1he infant is fussy or cries during
the physical examination by the doctor.

94.The infant accepts changes in solid
food feedings (type, amount, timing)
within 1 or 2 tries. ,

95.The infant moves much and for several

minutes or more when playing by self
(kicking, waving arms and bouncing).

Additio C ents

Frequently
5
almost l

never
almost l
never
almost 1
never
almost 1
never
almost 1
never

almost
always
6
2 3 4 S
2 5 4 5
2 5 4 5
2 3 4 5
2 5 4 5

almost
always

almost
always

almost
always

almost
always

almost
always

129

:53.» Unions av 42.9.9.th 0.3.9.30» benzene

 

...“...im... 2.259... ......

"...—5;.

3.1.2:... ...:

"ENE—T... c

 

 

 

a:

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

,. L912 2.. ...—sec. anemia—mo wins—EL: Ire—ma 2.5.: ....o ...—come 3:... 6.5.... .3333... no... AN ....Zcmmféncmw mu
_ . ..::..:: .3 5.3.5: .5: 1:35 P... .... 7:35.:— mfg—A12: moumoamoa 9:5... 93 miles—...-mm. ...:hcou .c._;r.mmo_.....o.1...3p...c.n
a vi .111... I.|Jvl...|.( Illul.‘ ; I: . |.|..A.. .‘. ..lllfil' I: EICWW
.. $ou
“ - -- - - :11... ,. «Q:
.N ...c
. - -- . -..- . we.
. Esw
m «who
m..ml.l....m...-l. a ...... .1. a .... so a... .. :27 .. e .N ... .3 ...“ ... ......s... ...a. ... . .... so : s. y. _ x s v .319. ...... e r as _ x we”.
M 3.;
l1: ...-.. . -I}. I - 1|- .
l .A. are: stems}.
mlflmkudzmsvssousws 373392373 sysawsusmisrmprmalinsofljaziuas
“obsmAlmfsTLJB epsuo_s_«m_.v:msu.owi .2 law ..Vsa.se..»::_ will...
i~3bofi se_.m...lfi_s.m._.m an spy:_F.~.l_.u..lJ.l.l_els.lp
“whim... .60-“..Malmw .... ....vafisoiu a labours... if: em.” 3‘s... rm slalmrmw.§
bulimia.......usfilw:u...$_~ ..vss 3...:le~.m_.v6.a_...mr.:s_..l 3.31
shrmfi.a:wlsusfi ......llafiul .4..r.lsoh.o-_.._.m~_a
f.anrush::ss?:Sussex: ..a £51..3.3.2.3531. aisles: 33.351. .6.
selenimsurmi 2.1.3.722“. wruiasulka lenses: 1.3.6 :Hliérlg Inflows :L
.-....»N .wnlumr...ikl..msu¢un_emisvsfisurms _ 3. along": ......vsom

Tenehufi ... $333.; ...».sz 2. , 63.21....insulin..._Hm:.:§s.fi.n.s33.1... A ...
Ghanigermfm13.7fm 3 C. e..sel_.3_..$rsurmsl_l www.msmllumevan
l _fivefmremovalmszrmWm?” n. ..mso_ls.g.vo lam. .......l_.......;
es 49: .5. t6 :2 a... m2 is ...s. 3...... as. as... :3 all 3. is a... .3
22.9.2»? 1.. Aprofiafinw—vomzmadua 16.96: X¢lmev7pxfi b..3€hw1§< succummc. blursfxsm. >+..>.+d<l.
.----.ii! -- .. (5:1--. 40

lust... a... as... t .....d 3...... to 5.2

:3.sz /.:.... .a.u.r.. -m.... ..tt: mqa<2201¢Han=GinmE(.flWO—tNLl LLZ<LzH

+15.

mean

-lS.

Easy

Diff

130

INFAHT TEMPERAMENT QUESTIONNAIRE - PROFILE SHEET
FOR 4 T0 8 MONTH OLD INFANTS
Revised 1977 by William 8. Carey, H. D., and Sean C. McDevitt, Ph.D.

Name of child Date of rating

 

Age at rating:_________ months days. Sex

 

Category score from Scoring Sheet:

 

 

 

Profile: Place mark in appropriate box below:

 

 

 

 

 

TActivity Rhythm. App/With. Adapt. Intens. Mood Persist. Distract. Thresh.
high arryth. withdr. slowly intense negative low per. lav dist. low
adapt.

Dl 4.96 3.05 3.05 2.61 4.13 3.48 3.85 2.84 4.55
4.40 2.36 2.27 2.02 3.42 2.81 3.03 2.23 3.79
3.83 1.68 1.50 1.42 2.71 2.13 2.20 1.63 3.04

very very
low rhyth. app. adapt. mild positive high per. high distL high

 

 

 

 

 

 

 

 

 

 

 

Diagnostic cluster:

 

 

 

rhyth. app. adapt. mild positive
arryth. withdr. ’ifigéz. intense negative
slovl
10" withdr. adapz. mild negative

 

 

 

 

 

 

 

 

 

 

 

Definition of diagnostic clusters used for individual scoring:

Easy - Scores greater than mean in no more than two of difficult/easy categories (rhythmicity,
approach, adaptability, intensity, and mood) and neither greater than one standard deviation.
Difficult - 4 or 5 scores greater than mean in difficult/easy categories (rhythmicity, ap-
proach, adaptability, intensity, and mood). These must include intensity and two scores must
be greater than one standard deviation.

Slov~to-varm-up - as defined above, but if either withdrawal or slow adaptability is greater
than one standard deviation, aetivity may vary up to 4.68 and mood may very down to 2.47.
Intermediate - all others. Intermediate high - b or 5 difficult/easy categories above the
mean with one greater than one standard deviation, or 2 or 3 above the mean with 2 or 3
greater than one standard deviation. Intermediate low - all other intermediates.

This child's diagnostic cluster Date of scoring
Comments: Scorer

 

 

LI ST OF REFERENCES

LIST OF REFERENCES

Ainsworth, M.D. Patterns of attachment behavior shown by the
infant in interaction with his mother. Merrill-Palmer
Quarterly, 196k, 10, 51 - 58.

American Psychological Association. Publication Manual-
Guidelines for nonsexist language in APA Journals.
Washington, D.C., 1977.

Bayley, N. Mental growth during the first three years: A
developmental study of 61 children by repeated tests.

Genetic Psychology Monoggaphs, 1933. 14, 1 - 92.

Bayley, N. and Schaeffer, E. Maternal Behavior and Personality
Development. In G. Medinnus (ed.) Readings in the Psy-
cholo of Parent Child Relations, New York: Wiley,
1957.

Bell, R.Q. Stimulus control of parent or caretaker behavior
2y offspring. Developmental Psychology, 1971, h,
3 - 72-

Bell, R.Q. History of the child's influence: Medieval to
modern times. In R. Bell and L. Harper (eds.), Child
Effects on Adults. New York: Halsted Press. 1977.

Bell, R.Q. Human infant - effects in the first year. In
R. Bell and L. Harper (eds.), Child Effects on Adults.
New York: Halsted Press. 1977.

Bell, S.M. and Ainsworth, M.D. Infant crying and maternal
responsiveness. Child Development, 1972, #3, 1171 -
1190.

Berger, M. Early experience with other environmental factors:
An overview. In H.J. Eysenck (ed.), Handbook of Abnor-
mal Psychology. San.Diego, Calif.: R.R. Knapp Pub.,
1973-

Birth, H.G. Functional effects of fetal malnutrition. Hos-
pital Practice, March 1971. 13h - 148.

Blank, M. The mother's role in infant development: A review.
In E. Rexford, L. Sander, and T. Shapiro (eds.), Infant
ngchiatry: _A1New §ynthesis. New Haven: Yale Univer-
sity Press, 1976

131

132

Bower, T.G.R. A Primer of Infant Development. San Francisco:
W.H. Freeman and Co., 1977.

Brazelton, T.B. Discussant's comments. In T.D. Tjossem (ed.),

Intervention Strategies for High Risk Infants and Youn
Children. Baltimore: University Park Press, 1976.

 

Briggs, D.C. Your Child's Self-Esteem. New York: Doubleday
and Co., 1975.

Butler, A.L. "Child's Right to Quality Day Care". Washington,
D.C.: Assoc. for Childhood Education International,
1970.

Caplan, G. Patterns of parental response to the crisis of
premature birth. Psychiatry, 1960, 23, 365 - 37h.

Carey, W.B. A simplified method of measuring infant tempera-
. ment. Journal of Pediatrics, 1970, 77, 188 - 194.

Carey, W.B., Lipton, W., and Myers, R. Temperament in adopted
and foster babies. Child Welfare, 1974, 6, 352 - 359.

Carey, W.B. and McDevitt, S.C. "Infant Temperament Question-
naire - Revised". 1977.

Chess, 8., Korn, S.. and Fernandez, P. Psychiatric Disorders
of Children wlth Congenital Rubella. New York: Brun-
ner7Mazel, 1971.

Chess, S. and Thomas. A. Differences in outcome with early
intervention in children with behavior disorders. In
M. Roff, L. Robins. and M. Pollack (eds.). Life History
Research in PsychopathologyI Vol. II. Minneapolis:
University of Minnesota Press. 1972.

Chess, 8.. Thomas. A., and Birch, H. Your Child is a Person.
New York: Penguin Books. 1976.

Church. J. Understanding YourChild from Birth to Three.
New York: Simon Schuster, 1976.

Clarke-Stewart, K.A. Interactions between mothers and their
young children: Characteristics and consequences.
Monographs of the Soclety for Research in Child Develop—
ment. 1973. 38, (6 - 7 Whole Number 153).

Combs, A. and Snygg, S. Individual Behavigg. New York:
Harper and Brothers. 1959.

Coopersmith, S. The Antecedents of Self-Esteem. San Fran-
cisco: WyH. Freeman and Co., 19 7.

133

Cox, S.H. "Family Background Effects on Personality Develop-
ment and Social Acceptance". Unpublished Doctoral
Dissertation, Texas Christian University, 1966.

Cramer, B. A mother's reactions to the birth of a premature
baby. In M. Klaus and J. Kennell (eds.), Maternal -
Infant Bonding. St. Louis: Mosby Co., 1976.

Cummings, S., Bayley, M., and Rie, H. Effects of the child's
deficiency on the mother: A study of mothers of men-
tally retarded, chronically ill and neurotic children.
American Journal of Orthopsychiatry, 1966, 36, 595 -
608 0

Department of Health, Education and Welfare - Children's
Bureau, "Care of New born Infants at risk in the hos-
pital". Washington, D.C. 1963.

Diggory, J., Klein, 5., and Cohen, N. Muscle action poten-
tials and estimated probability of success. Journal
of Expepimental Ps cholo , 1964, 68, 448 - 456.

Dodson, F. How to Parent. New York: New American Library,
1971.

Dunn, J. Distress and Comfort. Cambridge, Mass.: Harvard
University Press, 1977.

Ebeling, N. and Hill, D.A. Child AbugoIntervention and
Treatment. Acton, Mass.: Publishing Sciences Group,
InCo, 19750

Fitts, W.H. "Tennessee Self-Concept Scale Manual". Nash-
ville, Tenn.: Counselor Recordings and Tests, 1965.

Fitts, W.H. and Hamner, W.T. The Self-Concept and Delin-
guency. Nashville, Tenn.: Nashville Mental Health
Ctr. Pub., 1969.

Frankenburg, W. "Denver Developmental Screening Test:
Reference Manual". University of Colorado Medical
Center, 1975.

French, V. History of the child's influence. In R. Bell
and L. Harper (eds.), Child Effects on Adultg. New
York: Halsted Press, 1977.

Freud, S. Analysis terminable and interminable: Collected

papers, Volume 5, London, England: Hogarth Press,
1950. 31 - 357-

Fromm, E. Solfishness and Self Love, Psychiatry, 1939, 2,
507 - 523-

134

Gergen, K. The Concept of Self. New York: Holt, Rinehart,
and Winston, 1971.

Glass, G. and Stanley, J. Statistical Methods in Education
and Psychology. Englewood Cliffs, N.J.: Prentice-
Hall, Inc., 1970.

Gordon, I.F. Children's views of themselves. Washington:
Association for Childhood Education International, 1959.

Graham, P., Rutter, M., and George, S. Temperamental Charac-
teristics as predictors of Behavior disorders in chil-
dren. American Journal_of Orthopsychiatry, 1973, 43,

3. 328 - 339-

Gregg, G. Clinical experience with efforts to define indivi-
dual differences in temperament. In J. Westman (ed.),
lndividual Differences in Children. New York: Wiley,
1973-

Hamachek, D.E. Encounterswith the Self. New York: Holt,
Rinehart, and Winston, 1971.

Harrington, J. Violence: A clinical viewpoint. British
Medical Journal, 1972, 1, 228 - 231.

Hillestad, Mildred "Research: Process and Product". Service
Bulletin No. 1, Delta A Epsilon, St. Peter, Minn., 1972.

Irvin, N., Kennell, J., and Klaus, M. Caring for parents of
an infant with a con enital malformation. In M. Klaus
and J. Kennell (eds. , Maternal-Infant Bondiog. St.
Louis: Mosby, Co., 197 .

James, W. Ppinciples of Psychology. New York: Henry Holt
and Co., 1890.

Justice, B. and Duncan, D. Life crisis as a precursor to
child abuse. Public Health Reports, 1976, 91, 110 -
115.

Justice, B. and Justice, R. The Abusin Famil . New York:
Human Sciences Press, 1976.

Kagan, J. Resilience and Continuity in Psychological Develop-
ment. In A.M. Clarke and A.B. Clarke (eds.), Early
Eyperience: Myth and Evidence. New York: Free Press.

1977-

Kaplan, D. and Mason, E. Maternal reactions to premature
birth viewed as an acute emotional disorder. American
Journal of Orthopsychiatry, 1960, 30, 539 - 547.

135

Kiesler, S. and Baral, R. The search for a romantic partner:
The effects of self— esteem and physical attractiveness
on romantic behavior. In K. Gergen and D. Marlowe (eds. ),
Personality and Social Behavior, Reading, Mass.: Addi-
son Wesley, 1970.

Klaus, M.H. and Kennell, J.H. Mothers separated from their
new born infants. Podiatr. Clin. N.Am., 1970, 17,

Klaus, M. and Kennell, J. Maternal-Infant Bonding. St.
Louis: Mosby Co., 197 .

Kohlberg, L., LaCrosse, J. and Ricks, D. The predictability
of adult mental health from childhood behavior. In
B. B. Wolman (ed. ), Handbook of Child Psychopathology.
New York: McGraw-Hill, 1972.

Korner, A.F. Mother-child interaction: One or two way street?
Social Work, 1965, 10, 47 - 51.

Korner, A. F. Individual differences at birth: Implications
for early experience and later development. American
Journal of Orthopsychiatry, 1970,41, 4, 608 - 619.

Korner, A.F. and Grabstein, R. Visual alertness as related
to soothing in neonates: Implications for maternal
stimulation and early deprivation. Child Development,
1966, 37, 867 - 876.

Laing, R.D. Knots. New York: Pantheon Books, 1970.

Leiderman, P., Leifer, A., Barnett, C., and Williams, J.
Effects of'mother-infant separation on maternal attach-
meng behavior. Child Development, 1972, 43, 1203 -

121 .

Mahler, M. Thoughts about development and individuation.
Psychoanalytic Study of the Child, 1963, 18, 307.

Malone, C. E. The psychosocial characteristics of the chil-
dren from a developmental viewpoint. In E. Pavenstedt
(ed. ), The Drifters: Children of disorganized lower
class families. Boston: Little, Brown, 1967.

McBride, A.B. The Growth and 1Development of Mothers. New
York: Harper and Row Pub., 1973.

McDonald, R.I. The role of emotional factors in obstetric
complications: A review. Psychosomatic Medicine,
1968, 30, 222 - 237.

136

Medinnus, G. and Curtis, F. The relation between maternal
self-acceptance and child acceptance. Journal of Con—
sulting Psychology, 1963, 27, 542 - 544.

Morse, C., Sahler. L., and Friedman, S. A three year follow
up study of abused and neglected children. American
Journal of Diseases of Children, 1970, 120, 439 - 446.

Moss, H. Sex, age, and state as determinants of mother-infant
igteraction. Merrill-Palmer Quarterly, 1971, 13, 19 -
3 .

Moss, H., and Kagan, J. Birth to maturity: A study in
psychological deveflopment. New York: Wiley, 1962

Nie, N., Hull, C.H., Jenkins, J. G., Steinbrenner, K. and
Brent, D.H. Statistical Package for the Social Sciences.
New York: McGraw-Hill, 1975.

Prechtl, H. F. The mother-child interaction in babies with
minimal brain damage. In B. Foss (eds.), Determinants
of Infant Behavior, Vol. II. New York: Wiley, 1963.

Prugh, D. Emotional problems of the premature infant's
parents. Nurs. Outlook, 1953, 1, 461 - 464.

Purkey, W. The Self and Academic Achievement. Gainesville,
Fla.: Educational Research and Development Council,

1967.

Rheingold, H. L. The social and socializing infant. In D. A.
Goslin (ed.), Handbook of Socialization Theory and
Research. Chicago: Rand McNally, 1969, 779 - 796.

Rheingold, H.L. A comparative psychology of development.
In H. W. Stevenson, E. H. Hess, and H.L. Rheingold (eds. ),

Early Behavior Com arative and Developmental Approaches.
New York: Wiley, 1967.

 

 

Rheingold, J. C. The Mother Anxiety and Death. Boston:
Little, Brown and Co., 1967.

Robinson, J. and Shaver, P. Measures of Social Psydhological
Attltudes. Ann Arbor: University of Michigan Soc. Res.,
1973-

Robson, K. and Moss, H. Patterns and determinants of maternal
agtachment. Journal of Pediatrics, 1970, 77, 976 -
9 5-

Rosen, G.M. and Ross, A.0. Relationship of body image to
self-concept. Journal of Consultin Ps cholo , 1968,
32, 100.

137

The Volunteer Subject. New

Rosenthal, R. and Rosnow, R.
York: Wiley, 1975.
Predictions from

Rutter, M. Psychological development:
infancy. Journal of Child Psychiatry and Psychology,
19709 11. 9 " 20

Rutter, M. Epidemiological strategies and psychiatric con-
cepts in research on the vulnerable child. In E.J.
Anthony and C. Koupernik (eds.), The Chlld in his
Family: Children at Psychiatric Rigk. New York:
Wiley & Sons, 1974.

Rutter, M. Helping Troubled Children.

Pub., 1976.

Longitudinal data relating marriage satisfaction
Journal of Marriageand the Famil ,

New York: Plenum

Ryder, R.G.
and having a child.

1973. 35: 604 - 606.

Sameroff, A.J. "Infant risk factors in developmental deviancy".
Paper delivered at International Association for Child
Psychiatry and Allied Professions. Philadelphia,

July 1974.

Sameroff, A.J. and Chandler, M.J. Reproductive risk and the
continuum of caretaking causality. In F.D. Horowitz
(ed.), Review of Child Development Research, Vol. IV.

Chicago: University of Chicago Press, 1975.

Sameroff, A.J. and Kelly, P. "Socioeconomic status, racial,
and mental health factors in infant temperament" .
Unpublished manuscript reported in Thomas et al., (1977).

Perinatal characteristics of the

Sameroff, A.J. and Zax, M.
offspring of schizophrenic women. Journal of Nervous
191 - 1990

and Mental Diseases, 1973, 157,

Enhancing Self-Concept in Early Childhood.

Samuels, 8.0.
Human Sciences Press, 1977.

New York:

Scholom, A. "The relationship of infant and parent tempera-
ment to the prediction of child adjustment". Doctoral
dissertation, Michigan State University, 1975.

Sears, R.R., Maccoby, E.E., and Levin, H. Patterns of Child

Row, Peterson, 1957.

Rearing. Evanston, 111.:
A psychiatric study of parents

Steele, B.F. and Pollack, C.B.
who abuse infants and small children. In R. Helfer and
C. Kempe (eds.), The Battered Child. Chicago: Univer-

sity of Chicago Press, 1974.

138

Thomas, A., Chess, 8., Birch, H., Hertzig, M. and Korn, S.
thavioral Individualityln Early_Childhood. New York:
New York University Press, 1963.

Thomas, A., Chess, S. and Birch, H. Temperament and_Behavior
Disorders in Children. New York: New York University
Press, 19634

Thomas, A. and Chess, S. Differences in outcome with early
intervention in children with behavior disorders. In
M. Roff, L. Robins and M. Pollack (eds.), Life History
Reseapph in Psychopathology; Vol. II. Minneapolis:
University of Minnesota Press, 1972.

Thomas, A. and Chess, S. Temperament andlDevelopment. New
York: Brunner/Mazel, 1977.

Tocco, T.S. and Bridges, C. The relationship between the
self-concept of mothers and their children. Child

Study Journal, 1973, 3, 4.

Turner, R. The self-conception in social interaction. In
C. Gordon and K. Gergen (eds.), The Self in Social
Interaction. New York: Wiley, 19 .

Walters, C.E. Mothepelnfant Interaction. New York: Human
Sciences Press, 1976.

Warner, W.L. Social Class in Amgpica. New York: Harper
and Row, 1966.

Wells, L.E. and Maxwell, G. Self-Esteem: Its Concgptuali-.
zation and Measurement, Volume 20. Beverly Hills:
Sage Publications, 1976.

Wortis, R. The acceptance of the concept of the maternal
role by behavioral scientists: Its effects on women.
AmericgfisJournal of Orthopsychiappy, 1971, 41, 5
733 - 7 .

Wylie, R.C. The Self-Conc t Volume I: A review of metho-
dological congiderat1on gpdpeasuripginstruments.
Lincoln, Nebraska: UniVersity of Nebraska Press, 1974.

Yamamoto, K. (ed.) The Chlld and his Image: Self-Concept in
phe Early Years. Boston: Houghton Mifflin Co., 1972.

 

Yarrow, L.J. and Goodwin, M.S. Some conceptual issues in
the study of the mother/infant interaction. American

Joprnal of Orthopsychiappy, 1965, 35, 473 - 4 1.

 

"ICifijllflllflllllllfllllflr