{subhfi

 

This is to certify that the

thesis entitled
MEASUREMENT OF INFANT TEMPERAMENT

presented by

Howard Bonem

has been accepted towards fulfillment
of the requirements for

M.A. Psychology

degree in

 

 

~

 

{4?orp,
Date W‘ ‘23 437/ fiha? "‘5'

0-7639

MEASUREMENT OF INFANT TEMPERAMENT

BY

Howard Bonem

A THESIS
Submitted to
Michigan State University

in partial fulfillment of the requirements
for the degree of

MASTER OF ARTS

Department of Psychology

ABSTRACT
MEASUREMENT OF INFANT TEMPERAMENT
By

Howard Bonem

Attempting to develop a satisfactory measure of infant tempera-
ment, the present study analyzed the Carey Infant Temperament Scale;
concluding it is not a reliable measure. An alternate instrument, the
Michigan Infant Temperament Scale (M.I.T.S.) was devised. The M.I.T.S.
was administered to 160 mothers of infants 3 to 12 months old with
subsequent retest and inter-observer subsamples. Eight of 17 true-
false items achieved acceptable internal consistency. A ninth scale,
Persistence and Attention Span, did not meet appropriate internal con-
sistency criteria; the existence of more than one factor was offerred
as an explanation. Generally, inter-scale correlations were low to
moderate, while ratings of temporal stability and inter-observer
agreement were mostly moderate. Analysis of the M.I.T.S. with a larg-
er sample to cross-validate scale reliability and to examine construct
and scale validity was suggested. It was concluded that the present
instrument is quite adequate, but still in the experimental stages of

deve10pment.

ACKNOWLEDGMENTS

I would like to thank again the individuals who richly deserve
my thanks and appreciation for their help on this project.

This thesis has definitely been a group effort. The scale writing,
sample selection, data collection, key-punching, and coding was all
immensely aided by the many Psychology 490 students who worked with me.
Their investment and interest in the study was clearly out of proportion
to the minimal rewards they received. Without their help, the project
could not have been completed.

My committee stood by me with support and a faith (or expectation)
in my resolve to do a relevant and high-quality research project.

Bob Zucker, my chairman, provided a model of competence, dedication,
and thoroughness to the research and advisory process that I am sure will
stay with me throughout my professional career.

Helen Benedict and Neal Schmitt, who completed the thesis committee,
each contributed a special component to my progress. From my first day
at MSU, Helen has been a valued support and advocate, reinforcing me
with a belief (and expectation) in my ability to be a successful student.

Although my contacts with Neal Schmitt were not frequent, respect
for Neal's expertise and a desire to use him as a consultant and not a
statistics instructor spurred me to a preparation in test construction
and correlational techniques that I am only now beginning to fully value.

My wife, Debby, and our friends were of tremendous value. It was

11

their constant support, encouragement, and interest which helped me
move forward when I was feeling the weight and complexity of the task.
‘ Finally, the clerical and support staff of the Department of
Psychology deserve a special thanks for their aid in typing, duplicat-
ing, and collating of the protocol materials as does Sandy Harshman,

who typed this manuscript.

iii

TABLE OF

LIST OF TABLES . . . .
Chapter

I. Introduction . .

“The Concept of Temperament

Theory and Research .
The Factoral Approach

The Correlated Attribute Approach

A Proposed Integration .
‘Mbasurement of Temperament

Statement of the Problem

II o M8th0d 0 O o 0

Study I . . . .
Subjects . . .
Instruments . .
Procedures . . .

Study II . . .
Subjects . . .
Instruments . .

Procedures . . .
III. Results and Discussion

Study I . . .

Study II - Sample Characteristics

Reliability . .
Temporal Stability .
Validity . . .
Further Considerations

LIST OF REFERENCES . .
APPENDICES . . . . .

\Appendix A.
Sheet .

iv

CONTENTS

Carey Infant Temperament Survey

and

Scoring

Page
vi

p..-

nosapa
-§\Ot»\0<mtnto

26

26
26
26
28

28
28
29
30
32
32
32
35
44
46
52
54

57

S7

Appendix B. General Instructions, Departmental
Consent Form, Home Information
Sheet, M.I.T.S. questionnaire,
M.I.T.S. Supplementary Information
Form. . . . . . . . . .

‘Appendix C. General Information, Instructions
for M.I.T.S. questionnaire . . .

Appendix D. Letter to Fathers, General Information .

Appendix E. Letter to Caretaker, General Infor-
mation, Departmental Research Consent
Form 0 O I O I O O O 0

Appendix F. Items and their Loadings on each
Persistence Scale Factor, and Coef-
ficient Alpha and Item—total Correlations
for each Factor . . . . . .

Table 1.

Table 2.

Table 3.

Table 4.
Table 5.
Table 6.
Table 7.
Table 8.
Table 9.
Table 10.
Table 11.
Tableng.
Table 13.

Table 14.

Table 15.

LIST OF TABLES

Summary of Factor Analyses of the N.Y.L.S. Categories

The Conceptual Relationship of the Buss & Plomin Factors

(EASI-I) and the N.Y.L.S. Attributes
Carey Infant Temperament Scale Attribute Scales

Social Class Disorientation of Study II Sample and Two
and Eight-week Subsamples

Educational Levels of Tri-County Fathers and M.I.T.S.
Fathers

Scale Homogeneity and Item Heterogeneity of M.I.T.S.
Attribute Scales

Intercorrelations of Eight Attribute Scales of the
Michigan Infant Temperament Scale

Factor Analysis of Persistence and Attention Span
Scale

Two and Eight-week Test-retest Correlations of M.I.T.S.
Attribute Scales

Percentage of Fathers and Mothers Spending 0 to 6 or
more Hours per day with their Infants

Inter—observer Correlations for Eight Attribute
Scales of the M.I.T.S.

Comparisons of Father and Mother Ratings in Terms of
M.I.T.S. Scale Means and Correlation of Scale Differ-
ence Scores with Time Spent by Father with his Infant

Scoring Key for Carey Infant Temperament Scale

Scoring Key for Michigan Infant Temperament Scale
Items and their Loadings on each Persistence Scale

Factor, and Coefficient Alpha and Item-total corre-
lations for each Factor

vi

Page

18

33
34

36

38

40

41

45

47

49

51

65

83

95

INTRODUCTION

After many years of dormancy, the concept of temperament has begun

to re-emerge into psychological and personality theory. The return of

HM—

 

-——. __

this idea stems from the nature-nuture controversy--a debate over the

relative contributions of genetic—organismic versus environmental fact-

A__ -Av—u _._.—-'—‘

are to the development of human personality. For many years, behavior

m

was thought to be a result of body type, humoral influences, etc. The

 

concept of temperament easily fit into these theoretical constructions;
the idea of inherited temperally stable character traits and behavioral
styles are natural extensions of genetic-somatic-physiological explan-
ations (Sheldon, 1942). As the search for causal and explanatory con-
cepts switched to environmental-parental factors, the concept of temp-

ermant lost favor among personality theorists.. Specific behaviors as

M——
W. .—.'-
.-_.__-...a—-

-m‘

well as general personality traits were thought to deveIOp from inter-
actions with significant stimuli such as parents, traumatic events, etc.;
(Freud, 1948; Erikson, 1950; Sullivan, 1953). Inherited tendencies, if
existentjfiweTedviewed as only weak influences which might be easily
shaped or changed by environmental experiences. However, some theor-
ists continued to acknowledge the basic influence of temperament (Shirley,
1933; Gessell and Ames, 1937). Freud (1969) while committing himself to
exploring intrapsychic contributions to personality development accepted

the existence of genetic and somatic influences making use of "constitu-
tional" factors to help explain human behavior.

In the late 1950's temperament began to re-emerge with the review
of Diamond (1957), the writings of Allport (1961), and the work of

Thomas and his colleagues (1963). While Thomas' work is undoubtedly
l

heavily responsible for the renewal of interest in inherited styles of
functioning, their data collection methods (in-depth individual inter-
views with knowledgeable informants) have made it difficult for other
researchers to attempt to replicate and/or expand upon their basic find-
ings; interview and observational techniques are quite expensive, time
consuming, and inefficient ways of collecting data. Such methods limit
the range of investigation to an established few who are capable of
investing large quantities of time and money. Other approaches which
are more amenable to mass data collection techniques such as Q-sorts

and self-administered scales have attracted considerable attention (Buss
and Plomin, 1975; Carey, l970)--if for no other reason than the greater
ease and less expense of the investigative procedures.

Given such developments, the construction of a reliable and valid
infant temperament measuring instrument through which Thomas' efforts
might be furthered has become imperative. While there have been a few
attempts to accomplish this, an acceptable scale has yet to emerge./ It
is the aim of this study to attempt to build upon some of these earlier
efforts and to hapefully construct a reliable and valid instrument cap-
able of measuring infant temperament attributes.

THE CONCEPT OF TEMPERAMENT

Behavior may be viewed as consisting of two aspects: the content
of the actual sequence and the hay or way in which the instrumental act
is carried out (Thomas, Chess, Birth, Hertzig, & Horn, 1963). It is .
this latter "how" which the term temperament attempts to describe.
Bronson (1971) refers to this same characteristicaas the "style of life".
Such descriptors as the "underlying constant pattern" (Thomas, Chess,

Birch, & Hertzig, 1960) and the "formal characteristics" of behavior

(Thomas, Chess, & Birch, 1968) have also been used. Thus temperament,
in its broadest and most unelaborated form, refers to a set of under-
lying formal characteristics which influence or determine behavioral
style. These may include attributes such as speed of response, quality
of mood, regularity, and sociability.

Bronson (1971), who prefers the term "orientation", sets two basic
criteria by which the decision to include a personality attribute under
the rubric of temperament is made: temporal persistence and central-
ity. Temporal persistence refers to whether a characteristic remains
intact and unchanged over a period of time. Temperamental variables are
expected to be relatively enduring over the span of development. Cen-
tral variables are those which evidence generality, i.e., apply to a
wide range of behaviors in many situations. If a personality attribute
fulfills these criteria, it is then said to be a central orientation,
one which is central to the organism’s functioning and stable over time.

Buss and Plomin (1973) elaborate upon these criteria setting forth
three rules a temperamental characteristic must meet: 1) "The person-
ality disposition should have an adaptive value and therefore an evolu-
tionary history. Any inherited tendency must have survived natural
selection which means that it must have been useful in individual sur-
vival, breeding, and the rearing of the next generation..." Such attri-
butes would be central to the organism's style of functioning. 2) "The
personality dispositions should be present early in the life and show
some stability during childhood..." Although the environment will nat-
urally modify and shape the content of the individual's behavior, the
underlying style should, with some modifications, persist. 3) "There

should be evidence that the dispositions be inherited" (p. 513). If

temperamental characteristics are unlearned (Diamond, 1957), personal-
ity dispositions will at least partly be determined by genetic inher—

itance. This suggests that there should be some continuity in behav-

ioral style from generation to generation. However considerable

variation in temperament among family members and between generations

may occur due to genetic variations as well as such non-genetic influ
ences as birth complications, cerebral injury, etc.

These authors (Buss & Plomin, 1975) added two further criteria:
4) "If a personality disposition is inherited and shows at least mod-
erate stability during childhood, it should be present in adults" (p.
10). 5) "If a tendency has sufficient adaptive value to be passed on
through the genes, it is likely to be present not only in man but in
animals close to man" (p. 11). These latter two criteria,while of theor-
etical interest, do not appear to be essential to the definition and
classification of temperamental characteristics. While a tendency that
has persisted through generations of humans is likely to be present in
animals, an absence of this characteristic may be due to differences in
animal and human development, genetic characteristics, etc. Attributes
influential in human functioning may not have an adaptive value for
animals (and vice versa). In addition, a tendency once common to both I
animal and human, may, through many generations of environmental influ-
ence, have changed to such an extent that it does not readily relate
back to the original temperamental style.

The second addition-~presence in adults--is somewhat redundant
being an extension of criterion #2. Whether temperaments, which are
present at birth, continue into adulthood or slowly evolve (or mutate)

at some point in develOpment is an important theoretical question

subject to empirical investigation. To make this issue a prerequisite
for inclusion of a trait under the heading of temperament seems unneces-
sary and unproductive.

The above criteria provide a structure from.which infant person-
ality attributes may be divided into temperamental and learned compon-
ents of behavior. Only those traits which are present at birth, rel-
atively stable over time, serve (or once served) an adaptive purpose,
and have an inherited or genetic component can be classified as temper—
amental characteristics. Whether they are present in animals and in
human adults, although of theoretical importance, is not considered
vital. Our concern is directed at infants. Adequate investigation of
the continuity of temperamental traits from animal to human adult first
requires establishment of the existence and nature of the character-
istics in the human infant. Then, research into the effects of matur-
ity and environmental interchange become more meaningful and relevant.

The following is a review of those studies investigating the exist-
ence and nature of infant temperaments. The individual attributes dis-
cussed are assumed to fulfill the criteria listed above. This is but
an assumption and remains an important subject of investigation. How-
ever, examination of the research to date will indicate those variables
most commonly viewed as falling under the rubric of temperament. It is
these characteristics for which we h0pe to establish a measuring instru-
ment, and thereby re-evaluate our initial assumptions.

THEORY AND RESEARCH

Temperamental characteristics have been a frequent component in

personality theories ranging from Sheldon (1942) to Freud (1969). How-

ever, until the last twenty years, only a few empirical studies have

been attempted. Shirley (1933) conducted a longitudinal study of 0 to
24 month old infants. Using parental records, direct observation, and
physical measurements, she concluded that there "exists coherent pat-
terns of personality based on factors which are not "ironed-out" by
training, are often familial and are formed by age three months. One
variable, described as the tendency towards irritability or placidity,
appeared relatively stable over the two year period. Gessell and Ames
(1937) observed significant individual differences in human infants.
Their results suggested the existence of a social responsiveness trait
in infants.
The Factoral Approach

A major portion of current theory and research centers around three
or four temperamental factor structures. In a massive survey of over
100 biographies, Heymans (1908) hypothesizes three temperamental fact-
ors consisting of activity, emotionality, and primary versus secondary
function (the latter referring to the tendency to be influenced by
impressions of the moment versus a greater influence by residue of past
experiences). On a sample of 29 ninth and tenth grade adolescents, a
factor analysis of forty personality traits by Levandovskii (1970)
yields a similar set of factors: general activity, interpersonal rela—
tions, and self-control. Bronson (1971) also correlated a large number
of personality variables (34) and found seven bipolar variables to
account for most of the variance: reserved-expressive, somber-gay,
shy-socially easy, reactive-phlegmatic, explosive-calm, resistive-
compliant, and passive-domineering. The first three of the above traits
are grouped into a dimension called Expressiveness-Reserve while the

second three fall in the Placidity-Explosiveness dimension. A third

dimension consisting of the variable passive-domineering was also found.
However, Bronson reports this to be easily subject to organismic and
psychosocial influences. Bronson (1969) indicates that the first two
dimensions are relatively independent, unmodifiable, and enduring. Exam-
ination of their behavioral correlates does reveal one area of common-
ality: activity level.

Using an experimental procedure beginning at two or three days of
age, Birns, Barten, and Bridger (1969) found the following infant char-
acteristics as present at birth, and stable over the first four month
period: irritability, tension, and sensitivity. Activity level,
although showing some variance over time, also appeared to be a consist-
ent trait. The authors do not contend that the four traits are neces-
sarily independent.

Finally, Buss, Plomin, and Willerman (1973) and Buss and Plomin
(1975) who attempt an integrated theory of temperament, also arrive at
four factors: emotionality, activity, sociability and impulsivity.
While these four variables evidence factoral independence (low inter-
correlations) the authors have questions regarding the heritability of
the impulsivity trait. This stems from mixed data where the studies
are evenly divided between support and non-support for this hypothesis.

Allowing for "poetic license" (the vageries in labelling factors
derived from factor analysis) there appears to be some similarity in
the factors yielded by these studies. Activity level consistently
emerges as one of the three or four factors found by the above authors.
This is in spite of the variations in the Operationalizations of this
variable and the different methods used to measure it. The fact that

this factor emerged so clearly that "activity level" is almost

universally used as the descriptor provides additional support for the
existence of this trait.

Described as involving the expenditure of energy, activity level
may be evidenced as the amount, vigor, and vitality of a person's behav-
ior. It is very much how the individual acts and may refer to the
amplitude of a response as well as the tempo of a person's actions. As
such it tends to be influential in almost all behavioral occurrences.

An attempt to find a second common factor involves more problems.
Emotionality appears to be the most frequent descriptor used (Heymans,
1908; Buss and Plomin, 1975). This term does seem to include Bronson's
(1971) Dimension 1 (reserved-expressive, somber-gay, and shy-socially
easy) and may be similar to the sensitivity and tension characteristics
of Birns et a1. (1969).

Unlike activity level, a relatively unitary descriptor, emotional-
ity is used to represent more than one stylistic component of behavior.
Buss and Plomin (1975) define this trait in terms of arousal, reactivity,
and excitability. They suggest significant intercorrelations between
these three characteristics and treat them as aspects of a single fact-
or. Similarly, Bronson (1971) and Levandovskii (1970) view this cate-
gory as being composed of a number of interrelated variables. This is
in contrast to Thomas, Chess and Birch (1968) who identify three similar
attributes (intensity of reaction, quality of mood, and threshold of
responsiveness) as separate formal categories of behavior.

Although the above characteristics evidence enough similarities to
justify a grouping into common factors, this becomes a questionable pro-
cedure with the remaining traits. Heyman's (1908) primary versus sec-

ondary function, Buss and Plomin's (1975) impulsivity, Birns et al.'s

9

(1969) irritability, Levandovskii's (1970) self-control appear to have,
some commonalities. The atrributes of distractibility, attention span,
and persistence (Thomas et al., 1963) also fall into this category.
While no one term completely describes these characteristics, impulsiv-
ity (Buss & Plomin, 1975) does come the closest. As such, it is char-
acterized as involving the dimension of behavioral expression-inhibition,
as well as the variables of attention, persistence, response latencies,
etc.

The Correlated Attribute Approach

 

A second approach to the investigation and conceptualization of
temperament involves two major diversions from the factoral approach.
These differences, which have both methodological as well as conceptual
implications, center around: 1) How the temperamental categories are
formed; and 2) Assumptions regarding the statistical independence of the
temperamental traits. Whereas the factor studies started with either a
derivation of factors from other personality variables or an a priori
listing of hypothesized factors, the second approach begins with an
empirical analysis of infant behavior in an effort to extract tempera—
mental categories. These have been subject to further evaluation for
"fit" into the formal definition of temperament traits (see pages 3 to
6).

Secondly, no assumptions are made regarding the independence of the
attributes under study. The categories are seen as describing the
direct expressions of temperament. That is, they apply to the style by
which individual behaviors are effected as opposed to a more abstract
generalized description of behavioral style. The attributes are likely

to share significant interrelationships and thus not be factorally

10

independent. While this is theoretically and conceptually more cumber-
some, it can be more readily operationalized. Whereas temperamental
factors are seen as the "irreducable units" of temperament, this second

"units" .

approach primarily centers around the end product of these

As the New York Longitudinal Study (N.Y.L.S., Thomas et al., 1963;
Thomas et al., 1968) is the first in-depth empirical investigation of
temperamental characteristics, examination of the "trait" approach to
temperament shall begin with a fairly detailed description of their find-
ings.

The authors' interest in temperament evolved from a concern that the
organismic contribution to infant development had been ignored and neg-
1ected. Personality development is conceptualized as consisting of the
"interaction of a baby endowed with characteristics of initial reactiv-
ity and an environmental complex including familial and extra-familial
factors" (Thomas et al., 1963, p. ix). The authors set out to identify
those characteristics of individuality present in the first months of
life. The researchers then attempted to examine the degree of persist-
ence of these traits and how they influenced individual psychological
development.

Thomas et a1. performed a content analysis of twenty—two parent-
interview protocols which sampled the behavior of the informants' three
month old infants. Nine categories were found that scored continuously
and differentiated among individuals within each category. They were:
activity level, regularity, approach or withdrawal, adaptibility, intens-
ity of reaction, threshold of responsiveness, quality of mood, distract-

ibility, and attention span and persistence. Utilizing a three point

nominal scale, eighty infants were rated on the above categories.

11

Assessment was conducted by performing semi-structured open-ended
interview (approximately 1% hours in length) and then scoring the
interview protocol on the nine attributes. Stability among the cate-
gories over the first two years varied from a low of 27.5% for active
ity to a high of 92.5% in mood. The authors state that these variations
may be partly attributable to the develOpmental characteristics of the
individual temperamental categories. Each attribute may have its own
developmental sequence, with the process of individuation and differ-
entiation occurring at different times and rates in the life of the
infant.

Despite these variations, the temporal stability of Thomas et al.'s
temperamental ratings are quite good. Eight of the nine categories show
consistency over a two year period at the p‘<.001 level with only act-
ivity level evidencing any significant change. This only occurred in
the last six months of the testing period. These results, while derived
from one fairly homogeneous sample, strongly suggest the existence of-

a number of temperamental attributes.

Later research (Thomas et al., 1968) extended the investigation of
these temperamental attributes. Factor analysis of the nine categories
yielded a Factor A loading heavily in mood, intensity, approach/with-
drawal, and adaptability. This factor was found to be relatively con-
sistent over a five year time span. Scores on this factor were able to
differentiate normal and "clinical" groups of subjects. Two other fact-
oral trends also emerged from the analysis: a factor B is composed
primarily of threshold, rhythmicity, intensity, and adaptability; and a
factor C primarily loading in activity and intensity.

Attempts at replication of the N.Y.L.S. findings have been conducted

12

with mixed success. Graham, Rutter, and George (1973) rated a group of
British children on seven categories of behavior: mood, intensity of
emotional expression, activity, regularity, malleability, fastidious-
ness, and approach/withdrawal to new people. While some of these cate-
gories do differ from those of the N.Y.L.S. (malleability and fastidi-
ousness) and fail to include three other attributes (distractibility,
attention span and persistence, and threshold of response) their find-
ings provide some support for the N.Y.L.S. categories. The fact that
the categories could be reliably identified (inter-rater reliability =
80% for all except regularity, 64%) adds some validity to the hypoth-
esized existence of the five attributes common to both studies.
Inconsistencies in the above study as well as in the N.Y.L.S. led
Birns et a1. (1969) to the following criticisms of the parent interview
non-experimental approach: 1) Thomas et a1. rated their subjects at age
three months. Although the nine attributes found in these subjects are
viewed as being constitutional in nature, the extent to which behavior
may have been modified by environmental factors in the first three
months of life is unknown. 2) The data used in the N.Y.L.S. came from
maternal report. Given reliable maternal reporting, the ratings may be
based upon different stimulus conditions, rater judgements, etc. Birns
et a1. avoid these difficulties by rating the infants shortly after
birth. Subjects were exposed to a uniform set of stimuli in a neonatal
laboratory setting. The stimulus battery included presentation of a
moving picture, application of a cold disk to the infant's abdomen, and
other such procedures. Behavior was rated by three or four observers
who, along with the mother, were out of sight of the infant subject.

This solves the second criticism by exposing all children to the same

13

experimental conditions with their behavior being rated by neutral
judges.

The temperamental attributes being examined in this study were irri-
tability, soothability, activity level, alertness, vigor of response,
sensitivity, tension, and maturity level. At three or four months of
age, social responsiveness was added to the behavior ratings. It was
found that irritability, sensitivity, tension, and soothability showed
consistency from birth to four months. This suggests that certain
temperamental attributes are evident at birth and remain consistent
through age four months. The similarity of the variables being measured
provide support for the attributes used by the N.Y.L.S. The fact that
some of the traits were not found to be consistent may be as much attri-
butable to the rating procedures and the behaviors chosen to be rated as
to actual instability of the behavior characteristics. Problems in the
measurement of alertness and activity level, and in rating maturity level
are suggested as important contributing factors to the negative findings.
Such difficulties also highlight the problems of using a "controlled"
laboratory setting to evaluate infant temperament. Naturalistic observ-
ation may not only be more practical but also more sensitive to expres-
sions of temperamental characteristics.

A Proposed Integration

 

Discussion of temperament research has been conducted by dividing
the studies into two groups. The first was composed of those, who
through a variety of means, arrived at a three or four factor descrip-
tion of temperament. A major assumption was the relative independence
of the factors from one another. The second group included studies

which have attempted to investigate temperament through the

14

identification of attributes present at an early age and then traced
these behaviors for possible consistencies over time. Intercorrelation
of the attributes, rather than being avoided, was assumed. Emphasis

was upon uncovering the various components of unlearned behavioral
styles rather than a superstructure under which all the components might
be organized. It is suggested that Thomas et al.'s (1968) nine temper-
amental attributes and Buss and Plomin's (1975) four temperamental fact-
ors may both be valid ways of categorizing:temperament. Temperament may
consist of three or four independent dimensions, each consisting of a
number of related attributes. These attributes are capable of direct
translation into operational definitions, and therefore into behavioral
description. A dimension such as emotionality may be viewed as a group-
ing of attributes such as mood, intensity of response, etc. While the
dimensional approach might be more useful for research and theoretical
purposes, the attributes remains descriptive of an observable behavioral
style and may be more suited for clinical usage.

Support for this conceptualization is provided by Thomas et a1.
(1968), Scholom (1975) and McDevitt and Fox (1976). Each conducted
factor analyses of the nine N.Y.L.S. categories. The data for the form-
er were based on ratings of parental interviews while the two latter
studies utilized a questionnaire format (Carey, 1970). A listing of the
factors yielded by these three studies is found in Table 1. Examination
of the distribution of the attributes among the factors suggests a bas-
ic similarity. Certain attributes are consistently associated with each
other. Given different measuring and scoring techniques, as well as
differing subject samples (subjects were evaluated over a ten year per-

iod in three different parts of the United States) the achievement of

15

 

coo:

 

 

 

 

 

oaonwouna
huwfiwomufiuuH
huamsoucH mucoumamuom
mufimaouaH hufiaanuuomnuman hufiHHnfiuumuumfin
sua>auo¢ sua>auo< suw>auoa
powH> mwuoam o uouomm
ooaoumwmuom oaonmounH
suaumasusam suauwflawmm suaoaaausam
hufiowsnumnm mucoumfimsou m uouumm
suaaapmuamua suaaaawuamn<
Hwamuonu«3\nuwouaa< Hmsouosua3\nuoouoo< Hmsouonua3\somouaq<
suaaapmuamca afloammuas summaouam
suaaanauomuumaa coo: woos
suaaapmauom woo: a genome
xom a uuw>oaoz aoaonom .Hm um uneasy

 

moauowoumu .m.A.w.z on» we mouhaoq< nouoom mo mumasom

H magma

l6

partial replicability can be interpreted as reflecting some basic com-
monalities.

When Scholom's three factors are compared with those previously
discussed (see pp. 7-10) commonalities do emerge.1 Certainly the auth;
ors' choice of terms: Energy-Activity, Impulsivity-Consistency, and
(to a lesser extent) Mood-Emotionality and Sociobility suggest the
description of a similar group of traits. In fact, the formal defini-
tions of the terms that are supplied by the authors are also quite close.
Buss and Plomin (1975, p. 30) describe "activity level" as dealing with
energy output while Scholom (1975) also refers to "energy" as the level
or amount of expended energy. The same relationship is seen in "impuls-
ivity", defined as dealing with the dimension of control as well as
such variables as decision times, boredom, and persistence. It appears
to have some relationship to "consistency" which is described in terms
of constant versus cyclic styles of behavior. Scholom's "mood" seems to
encompass both "emotionality" and "sociability". It is defined as)
approach, adaptivility, mood, and threshold and can be compared to arous-
al, reactivity, and excitability (Emotionality) and preference for and
response to other people (Sociobility).

Examination of the content of the Buss and Plomin factors indicates
an omission of some of the N.Y.L.S. attributes and a duplication of

others. This author rated the items of the EASI-I Temperament Survey

1Scholom's results are probably the most complete and clear of the
three studies; the second and third factors of Thomas et a1. were not
clearly delineated while McDevitt and Fox failed to include the
attribute "persistence" in their analysis. For purpose of the dis-
cussion, Scholom's data will be considered representative of all
three studies. '

17

(Buss & Plomin, 1975) according to the N.Y.L.S. rating system.2 Each
item was scored for the attribute category it most resembled. Table
2 lists the result of this procedure. It can be seen that, with some
overlap, seven of the nine N.Y.L.S. attributes score on the EASI-I
categories. The two omissions are regularity and adaptability. While
there are definite differences in the two systems, it does appear that
the Buss and Plomin factors are readily translatable into the N.Y.L.S.
nine attribute categories. This is added support for the proportion
regarding the complementarity of these two descriptive systems.

Finally, if one assumes that these two approaches are both describ-
ing a similar if not the same phenomenon, i.e., styles of behavior that
are adaptive, inherited, and stable over time, then the differences may
be no more than different ways of cutting the same pie. There is the
more persimonious though somewhat behaviorally removed factorial organ-
ization versus a collection of attributes that are by definition
directly related to observable behavior but possessing unknown inter-
relationships.

In the final analysis, the question of the validity of either theory
y or their comparability becomes a basic issue of the utility or power
of the approach for the particular research or clinical question under
study. While factors may prove to be conceptually more useful, the

N.Y.L.S. attributes yield more direct and detailed information regarding

2Although the Buss and Plomin factors are used for illustrative purposes,
these comments also apply to the other factoral approaches to temper-
ament. The EASI-I was chosen due to the clarity and comprehensiveness
of the Buss and Plomin system.

18

Table 2

The Conceptual Relationship of the
Buss & Plomin Factors (EASI-I)
and the N.Y.L.S. Attributes

 

 

Buss & Plomin Factors N.Y.L.S. Dimensions

Emotionality Threshold
Mood
(Intensity)*

Sociability Approach/Withdrawal
Mood

Activity Activity Level
Intensity

Impulsivity Attention Span & Persistence
Threshold
CDistractibility)

*An attribute listed within parentheses scored marginally on the
factor.

19

subject behavior. As such they may be of greater clinical utility.

Measurement of Temperament

 

Most efforts to investigate temperament have utilized a semi-
structured interview to gather information (Thomas et al., 1963; Thomas.
et al., 1968; Graham.et al., 1973). Birns et a1. (1969) go to the other
extreme, attempting to evaluate temperamental characteristics in the lab—
oratory. Although this method allows the experimenter to control the
stimulus conditions and directly observe the behavioral responses, such
a procedure typically does not easily translate to field research or
clinical applications because the number of situations observed is lime
ited.

The interview approach is best typified by the N.Y.L.S. Thomas et
al. gathered data from.semi-structured interviews with mothers of the
infants under study. The one to two hour interviews were tape recorded
and later replayed for scoring. This procedure is not only quite time
consuming but also raises questions regarding reliability of the inter-
viewer, data source (mother), and the rater. Thomas et al. found high
inter-judge and intra-scorer agreement (90%). While the authors appeared
to achieve a satisfactory level of reliability, whether this could be
maintained across studies by different investigators and raters is ques-
tionable. The extensive training and years of experience available to
Thomas and his colleagues is not available to others. This potential
lack of consistency in scoring interview data may be partially respons-
ible for some of the inconsistencies found in the N.Y.L.S. and other
similar studies (Graham et al., 1973; Birns et al., 1969).

Infants cannot be directly interviewed. Thus a knowledgeable

observer must serve as a reporter of these subjects' behavior. Although

20

a mother (or father) may be a most knowledgeable source, the accuracy
and consistency of their reporting remains questionable. There is no
way to completely objectify parental ratings. However, the N.Y.L.S.
procedure involving Open-ended unstructured interviews appears to add
extra "noise" to a potentially unreliable technique. Although the inter-
viewer(s) may be highly trained, it is likely that there will be vari-
ations in the rapport established with the maternal reporter, the types
and amount of stimulus questions needed to elicit the required infor—
mation, and the clarity of the reporter's responses. A highly struc-
tured interview schedule or questionnaire will not increase the parent's
objectivity in relation to their infant's behavior. However, it is a
more consistent stimulus, relatively free of interviewer differences,
clarity of parental responses, etc. A questionnaire would also be much
shorter and more readily available for clinical applications.

Attempts to gather temperamental data through a questionnaire have
been made by Buss and Plomin (1975), Carey (1970), Scarr and Salapatek
(1970), Peterson, Anderson, and Cain (1976) and Feiring and Taylor
(1976). Buss and Plomin devised a 20 item scale (EASI-I) containing
four subscales, each measuring one of the four temperaments under study
(emotionality, impulsivity, sociability, and activity). A revision of
the scale (EASI-II) resulted in a scale of high reliability (.83) and
factoral consistency.

In the N.Y.L.S., reliance was exclusively placed upon direct obser-
vation and parental report. Carey (1970), feeling the need for a quick-
er, standardized method of obtaining temperamental information of
infants in his pediatric practice, devised the Carey Infant Temperament

Scale (C.I.T.S.). Seventy statements were constructed, each allowing

21

three possible responses. The individual items generally described
situation-specific behaviors. For example, item 42: (a) If bath by
different person or in different place, readily accepts change first

or second time, (b) May or may not accept, (c) Objects consistently to
such changes. The parent/rater is requested to only report actual
infant responses and may omit an item if the three behavioral choices
are not suitable. Seventy-six ratings are derived from the seventy
items (six items are scored on two scales) yielding nine average react-
ivity scores (see Appendix I for questionnaire and detailed scoring pro-
cedures).

Statistical evaluation of the C.I.T.S. has not been extensive. The
individual items appear to have good face validity-~the items seem to
directly relate to the scale under which they have been included. How-
ever, Carey (1976) reports internal consistency for a 200 subject test
administration of only 0.76 for the total instrument. McDevitt and Fox
(1976) report alpha coefficients for internal consistency on the individ-
ual scales as ranging from .27 to .72 with a median of .47. Their find-
ings raise questions concerning the interrelationships of the particular
C.I.T.S. items. There may be some extraneous factor(s) which account
for the high total test alpha. The appropriateness and placement of the
individual items definitely needs further evaluation. Additionally, the
low number of items per scale may also contribute to the low reliability
scores.

Temporal stability of the questionnaire uas initially evaluated by
conducting a retesting of four subjects, two weeks after their initial
rating (Carey, 1970). "...there was agreement without exception in the

five major categories for these three who were all between 6 and 7

22

months of age" (p. 191). A later test—retest reliability evaluation of
20 subjects seated two weeks apart yielded a range from 0.72 for per-
sistence to 0.94 for rhythmicity. A correlation of 0.84 is reported for
the whole instrument (Carey, 1976). This second reliability test allows
some confidence in the instrument's short-term temporal stability. How-
ever, the range of scores for the individual scales is somewhat large
and further examination of these scales should be conducted.

The existence of temperamental attributes remains a subject of
empirical investigation. As such, the issue of the external validity
of a temperament scale becomes one of hypothesis testing as well as
scale validation. To date, validation efforts have centered around var-
ious comparisons of the C.I.T.S. with results obtained by the N.Y.L.S.
This includes comparing the mean scores of 101 subjects (age 3% to 8%
months) with normative data collected by Thomas et a1. (1968). Although
there are some differences in the scores, for example, activity yielded
a mean score of 0.52 $10.32 on the C.I.T.S. versus 0.80 in the N.Y.L.S.
interview (Period 1). Carey (1970) suggests that these differences are
of no practical importance.

When the same criteria for the determination of "difficult children"
is used, both techniques yield similar percentages of this category (10%
on the C.I.T.S. and 8% for the N.Y.L.S.). Finally, Carey reports that
four infants were evaluated both by the C.I.T.S. and by an interviewer
familiar with the N.Y.L.S. interview technique. For three of the four
children, there was a high level of agreement.

A number of similar measurement scales have also been constructed.
Fairing and Taylor (1976) revised the C.I.T.S. using only five attribute

categories (rhythmicity, adaptability, approach, intensity, and mood),

23

eliminating the six dual—scored items, and moving to a three-point
numeric scoring system (0, 1, 2). Two forms of this instrument were
used in their study of infant and secondary parent influence upon
maternal behavior: Form I for infants age 2-0 to 7-30 months and Form
11 for children 8-0 to 19-30 months. No validity or reliability data
are reported.

Similarly, Scarr and Salapatek (1970) also constructed a nine
scale temperament survey. They report internal consistency among the
nine scales as having median correlations ranging from .15 to .55 and
a median item-total correlation of .25.

A second approach is reported by Pederson, Anderson, and Cain, Jr.
(1976). Reviewing the above scales, the authors suggest that these
instruments all seem vulnerable to response sets such as social desir-
ability or acquiescent tendencies. In an attempt at further refinement
of the N.Y.L.S. approach, the authors designed the Perception of Baby
Temperament Scale (PBT). This instrument has nine attribute scales with
each scale containing six items. Administration is in a Q-sort format
with statements appearing on 3x5 cards. Items are sorted into three
response categories: "Very much like my baby"; Sometimes or occasional-
ly like my baby"; and "Not at all like my baby". Data on 26 families
resulted in corrected split—half reliabilities for five of the scales
(activity, rhythmicity, adaptability, approach, and mood) ranging from
.54 to .69. Reliabilities of the other four scales are not reported.
Partly attributing the moderateness of the scores to the low number of
items per scale, the authors view the PBT as a potentially useful instru-
ment. However, the use of the Q-sort format may prove to be a hindrance

in large-scale applications.

24

Given the above data regarding reliability and stability, the
results of studies employing the C.I.T.S. must be viewed with consid-
erable caution (Carey, 1970, 1972a, 1972b; Carey, Lipton, & Myers,

1974; Carey, 1974; Scholom, 1975). Although Carey (1976) concludes that

the C.I.T.S. is "more-or—less accurate,"

the scale contains many weak-
nesses. Test-retest reliability is within acceptable limits. However,
a determination of both the accuracy and construct validity of the mea-
suring instrument has yet to be clearly established. The C.I.T.S. seems
to tap similar material as that investigated by the N.Y.L.S., but it is
unknown whether difference in the results of these two studies is due to
variations in the subject sample or measurement discrepancies. Addi-
tional problems involve: 1) scoring of some items on two attribute
scales; and 2) the use of a three point scoring system. Scholom (1975)
suggests that a utilization of a 5 or 7 point scale might eliminate some
scoring bias through greater differentiation of response options, and
allow assessment of the relative magnitude of the effects of the vari—
ables by providing more within item variability. 0n the other hand,
there is the question of whether a three, five, or seven point scaling
is appropriate. The respondent's ability to accurately judge "varia—
bility" in their infant's behavior remains an untested assumption. A
change to a different format that avoids such difficult discriminations

may be more apprOpriate.

Statement of the Problem

 

Further investigation of infant temperament depends upon accurate
statistically valid systems of measurement. While the N.Y.L.S. has
gained wide popularity in both the literature and in clinical practice,

there is no established method for measuring the nine attribute

Ill .tlilll .‘(llxfi illutbr.

l,‘.l,

(till

25

constructs. Though the C.I.T.S. has the potential for filling this gap,
at present it is of questionable utility. The purpose of the present
study is to evaluate, revise, and statistically validate the C.I.T.S.
with goal of establishing an instrument capable of being used for
researching the temperament concepts of Thomas et a1. (1963). The pres-
ent study is divided into two parts: 1) an evaluation of the present
version of the C.I.T.S. Special attention will be given to the issues
of item relatedness and item variability, length of the individual
attribute scales, and the sensitivity of the three point response scale
2) revisions will be made in response to any weakness uncovered during
the above evaluation. The revised instrument will then be tested for

temporal stability, internal consistency and external validity.

METHOD

Study 1
Subjects

The sample for Study 1 consisted of 180 parents of children, two to
three years of age, who have been enrolled in day care centers and nurs—
ery schools in and around the Michigan State University community. The
families were largely university affiliated (students and faculty), and
from a fairly high socio—economic background. All were subjects in
Scholom's (1975) study of infant and parent temperament as possible pre-
dictors of child adjustment. Initially, 395 families were contacted for
participation in the study, and 292 agreed to fill out the forms with 180
ultimately returning completed questionnaires.

Instruments

 

Study I utilized the Carey Infant Temperament Scale, a 70 item
forced choice instrument which yields scores on the nine temperament
categories found in the New York Longitudinal Study (see Appendix A).

The following is the description of the categories used to assess
infant temperament as given by Scholom (1975):

1. Activity Level

The motor component present in a given child's functioning, and
the diurnal proportion of active and inactive periods, plus protocol
data on mobility during bathing, eating, playing, dressing, and hand-
ling, as well as information concerning the sleepdwake cycle, reaching,
crawling, and walking were used in scoring the category.

2. Regularity

The predictability and/or the unpredictability in time of any
function was analyzed in relation to the sleep-wake cycle, hunger,

26

27

feeding pattern, and elimination schedule.

3. Approach or withdrawal

The nature of the response to a new stimulus, be it a new food,
new toy, or new person, provided information relevant to this category.

4. Adaptability

Responses to new or altered situations. One is not concerned with
the nature of the initial responses, but with the frequency with which
they were successfully modified in desired directions.

5. Intensity of reaction

The energy level of response, irrespective of its quality or direc-
tion.

6. Threshold of responsiveness

The intensity level of stimulation that was necessary to evoke a
discernible response, irrespective of the specific form that the response
might take or the sensory modality affected. The behaviors utilized were
those concerning reactions to sensory stimuli, environmental objects, and
social contact.

7. Quality of mood

The amount of pleasant, joyful, and friendly behavior, as contrasted
with unpleasant, crying, and unfriendly behavior.

8. Distractibility I

The effectiveness of extraneous environmental stimuli in interfer-
ing with, or in altering the direction of, the ongoing behavior.

9. Attention span and persistence

These two categories are related. Attention span is the length of
time a particular activity is pursued by the child. Persistence refers

to the continuation of an activity direction (PP. 28-30).

28

The number of items in the C.I.T.S. categories varies from five to
sixteen. The items are scored on a three point scale with one being
assigned to responses reflecting positive, high intensity, or adaptive
behaviors and a three to negative, low intensity or unadaptive behav-
iors.

Procedures

A sample of completed C.I.T.S. forms was obtained by utilizing all
the available data of Scholom (1975). He first contacted parents by
letter. This was followed by a phone call to make more personalized
contact and to answer questions about the study. The C.I.T.S. was then
mailed to these parents as part of an infant and parent temperament bat-
tery .

Parents were instructed to answer the C.I.T.S. based on their child
as an infant (2 to 3 years in the past). Although this procedure runs
the risk of faulty parent memory and halo effects, given that the prim-
ary goal of this study was to examine the internal consistency of the
scale (and not to either examine issues of validity or to obtain temper-
ament ratings for hypothesis testing) the effects of this data collection
method should, if anything, inflate both internal consistencies and

inter-scale correlations.

Study II
Subjects

The Study II sample consisted of 160 three to twelve month old
infants and their parents. Names of potential subjects were obtained by
sampling at regular intervals from birth certificates of children in the
specified age range that are on public file at the local county clerk's

office. Names and addresses of the parents were recorded. Each mother

29

was then contacted by phone, given a brief description of the study,
and asked if she would be willing to participate.

Initially, 205 mothers were contacted. Of this number, 192 agreed
to participate, and 160 ultimately returned data (83.3%). The sample
contained 51% female and 49% male infants. The average age of the
infants was 224 days or 34.5 weeks (SD=85.9 days).

Other relevant characteristics of the sample include: average
birth weight of the infant, R97.42 pounds (SD=1.66); number of children
in the family, E§1.80 (SD-1.12); percentage of children receiving care
by a day care center, 1.2; by an "alternate caretaker", 30.4. The data
suggest that children in the sample had a normal birth weight, are on the
average the younger of two children, do not attend day care centers, but
may be taken care of for more than three hours per week by someone other
than an immediate family member.

Instruments

 

The Midhigan Infant Temperament Scale (M.I.T.S.) was developed to
assess infant temperament as part of Study 11. The M.I.T.S. is a revised
and elaborated version of the C.I.T.S. To achieve an acceptable level
of internal consistency on the attribute scales, new items were written
and old ones revised so that each scale consisted of 25 to 30 true-false
items with the complete instrument totaling 243 items. The change to a
true-false format (as opposed to a three point response scale) was made
in an attempt to provide test respondents with as simple and unambiguous
response Options as possible. The items were balanced to control for
possible acquiescence sets. The items were also randomly ordered to
avoid possible response sets arising from the organization of sample

infant behaviors around content areas (e.g., bathing, feeding, dressing).

30

Procedures

 

The M.I.T.S. was administered in a way similar to Scholom's proced-
ure. Contacted by phone, the study was briefly explained to the parents
and answers to any questions were provided. Feedback on questionnaire
results was offered to interested parents. Once consent was obtained,
the M.I.T.S. was mailed; the parent was asked to complete it and accom-
panying demographic information sheets. Protocols were returned in pre-
stamped envelopes (see Appendix B for sample protocols and instruction
forms).

Two subsamples of the above parents were recontacted shortly before
either a two (21860) or eight (22'60) week interlude. They were again
asked to complete the M.I.T.S. The parents were reassured that this

N

retesting was to be ...for us to see whether your infant's behavior has

" Forms were sent

changed in any way, or remained pretty much the same...
to the mothers three to five days before the two or eight week period
lapsed (see Appendix C for parent letter and instructions). Twenty-seven
parents in the two week retest sample (45%) and 32 mothers in the eight
week group (53.3%) returned completed questionnaires. The actual inter-
val between administrations for the two and eight week retest groups

was 21.6 days and 59.3 days, respectively.

In order to begin an assessment of inter-observer agreement and
scale validity, two additional subsamples were also obtained: 1) 100
fathers were sent the M.I.T.S. They were told that although they were
under no obligation to participate in the study, it was important to see
how fathers judge their infant's behavior as well as mothers (Appendix D.

contains the father's letter and instructions); 2) Twenty-six parents of

children who either attended a day care facility or were cared for by

31

someone other than an immediate family member were asked for permission

to administer the M.I.T.S. to a teacher or caretaker familiar with their
infant. These individuals were sent a descriptive letter, instructions,
a copy of the parent's permission form (page 2 of Supplementary Form),

a consent form for themselves and the M.I.T.S. (see Appendix E). Given

parent permission, they were offerred feedback on the scores obtained on
the instrument. Thirty-seven fathers returned completed questionnaires

(37%) as did 14 caretakers (53.8%).'

RESULTS AND DISCUSSION
Study I

Reliability characteristics of the nine C.I.T.S. attribute scales
are given in Table 3. The coefficient alphas, a measure of item relat-
edness (internal consistency) range from .30 to .68. There is a great
range in the corrected itemrtotal correlations. This indicates that
the internal consistency of the C.I.T.S. attribute scales is not high.
Variables other than those targeted for measurement are likely to have
had a large influence upon scale scores. 0n the basis of this analysis,
it appears that the Carey instrument does not reliably measure the nine
temperamental variables.

Study II — Sample Characteristics

Study II involved the administration of the M.I.T.S. to a sample of
the local pOpulation. The study employed a sampling of subjects by
regular intervals of age and had a quite adequate, but less than complete
return rate. Before analyzing the individual attribute scales, it is
important to first look at the characteristics of the final sample and
to assess its representativeness in terms of the mid-Michigan pOpulation
of infants and their families.

The social class distribution of the sample is presented in Table 4.
Social class was calculated using Hollingshead's Two Factor Index of
Social Position (1957). The data indicate that all social classes are
represented with social classes 2, 3, and 4 representing 75.7% of the
sample. Thus the sample appears to be fairly heterogenous with respect
to social class.

Even so, an analysis of the sample's representativeness in compar-
ison with the Tri-County population (Ingham, Eaton, and Clinton

32

 

 

Hm. I so. has. n moguamamumm

 

 

 

 

Ha. I no. Hon. .5. hufiafinmuomuuoan
mm. I so. awe. ca moo:
AN. I mo. HHS. NH suamemuem
an. I mo. «cc. 5 vHonmoush
QN. I mo. coo. NH suaaanauame<
NN. I oo. mac. 0 accounm<
mm. I mo. nmm. m hufiowasuhsm
cw. I no. mow. m >ufi>wuu<
coaumaouuoo HmuOuIaouH madam usoaoammooo maouH mo Honasz mmmmwmmm¢.
mo owsmm

 

 

Aowauzv mmamum muonauuu< oamum ucoamuoaaoa usmwsH moumu

-m «Home

. .I ~ t.”-
I 0 II .Illw.,1ul‘ '. ‘
A

«an . .v .53.. m .1,
.. . .. . 19, f at:
( 1...... ail. 5:? scarab... , .. 1?. O .

34

.m.z “camp mm.au x “mum so maaamm ummumu xmoz unmfim mam mHaBMm mama mo momwummaoo

N

o

.m.z “same mm.mu x mmmm so mamamm umwomu xmoa oBu mam masses same we somfiumoaoom

N

 

 

 

&H.m Nn.m No.5

NH.wN Nm.mm NH.©N

Nq.¢m Nm.mm Nw.Hm

Nc.m Nm.wH Nw.na

Nw.mH Nm.mH Nowoa
AmmuzvnxmmS unwwm AnmlzvmxmoB 039 AooHHZV onEmm cfimz

Amumuonms emfiaachae >

Amumuonms emaaaxm
ecu emHHmeHammv >H

Aamowuofio
.mamaowmmmmoum HocHZV HHH

Afloacomuoa
mmocamsn Hm>ma noon:
.mamcoammmwoua someway HH

Amo>au=omxm
.mHchfiwmmmouo
acnmzv H

 

mmmao Hmfioom

mmamammosm xooBIustm one 039 one oaaamm HH hooum mo coausnfiuumwo mmmao HmHoom

s «Hams

35

counties) is desirable. A rough comparison of the sample and population
in terms of educational level can be made by combining the educational
categories of Hollingshead (1957) into four groups: less than four
years of high school, high school completion, some college, and four
or more years of college. Table 5 lists percentages for each grouping
in the M.I.T.S. sample and that of fathers under the age of 35 who live
in the Tri-County area (U.S. Bureau of the Census, 1972). It can be
seen that the percentage of fathers with at least some college education
is much higher than that of the Tri-County's father population. Thus
the fathers of the infants in our study are likely to be more highly
educated than the average father in the Tri-County area. This is prob-
ably due to the greater likelihood of families with higher levels of
education to be interested in and willing to fill out a questionnaire
on infant behavior.

Analysis of the two test-retest subsamples for representativeness
of the main sample in terms of SES level was also done (see Table 4).
These data indicate that the distribution of SES in the fathers of the
two and eight-week groups resembles the main sample. Thus the retest
groups may be considered as representative of the overall Study II
sampling, allowing the use of these samples for analysis of temporal
stability.
Reliability

Each of the nine attribute scales of the M.I.T.S. were subjected to
reliability analysis. This involved the calculation of coefficient
alpha and item-total correlations, and a factor analysis of any scale
that appeared to be heterogeneous with respect to infant behavior and

temperament. For example, persistence, which had a low coefficient

36

 

 

A

 

moo. vm.mmuma ~.am nmx
.m
nma I n
. .m
omm ma I m
.owo mo munch o coco
mmoH mum souoaaso can own mo mums» mm woos: ma oaonomso: 0:» mo
one; onu noes: ca mmfiHfiamw omaalosmomsn now who mouswwm coaumHsoom .ouoz
owoaaou mo
m.mq N.mH mums» ouoa no noon
mm.a~ m.aa owmsaou maom
~.q~ H.Nq Hoonom swam
Hoozom swam mo
H.n m.HN mumm> “now can» mama
mamamm aw unmoumm maowumaomom cw unmoumm cofiumosom mo Ho>mq

muosumm .m.H.H.z was mumsumm SassooIHuH mo

m «Hams

mHm>oA Hmco«umo=om

37

alpha (.290) and many items with low item-total correlations, was sub-
jected to a factor analysis.

The goal of these analyses was to construct, from the pool of items,
nine internally consistent scales of infant temperament. A coefficient
alpha of .70 was set as the minimum acceptable value for internal con-
sistency. This level was chosen so that measurement error would be kept
at or below the 50 percent level. In addition, a .20 corrected item-
total correlation was considered the minimal level acceptable for
inclusion in a final form of a scale.3 After the scales were revised
to meet these criteria, the number of items per attribute scale was set
at 17. This number was chosen as the lowest number of items that could
fulfill the following criteria: 1) achieve a minimum .70 coefficient
alpha for each scale; 2) result in a revised instrument that could be
completed in 30 minutes or less; and 3) that would sample a complete
range of behaviors that are theoretically influenced by the proposed‘
temperamental variables.

It proved possible to meet the above criteria for eight of the nine
temperamental attribute scales. Coefficient alpha for these eight
scales ranged from .705 to .859 with a median of .737. Table 6 con-
tains the coefficient alpha and the range of corrected item-total cor-
relations for these eight scales. The ninth scale, persistence, was
deleted due to lack of item homogeneity.

Inter-correlations of the eight adequate attribute scales was done

3This criterion was at times relaxed in order to insure complete samp-
ling of an attribute. No more than two items per scale were permitted
to be below the .20 level.

’

62-»

.mmamum oaom pom

so.

an.

mm.

mm.

0N.

mm.

am.

He.

.u chaos:

.30H wagon maoua can no one afico sows momma: none as maoauma
Iouuou onu mo husmosou Hmuusmo onu .mH. Baden on pace mosmuxm swamp mnu swoonuam

mm. mm.
cc. AH.
om. ma.
Ne. mo.
no. «a.
No. NH.
me. ma.
00. AN.
mmmmm

 

mmcowumaouuou HmuouIaouH wouoouuou

em.

as.

on.

mm.

mm.

an.

on.

om.

 

mama< semaoaemmou

.mamom sumo mo mama“ ma onu How momcmu who ooumwa chfiumHouuoom

suauaanusam
suaaaaauomuumaa
nomouoo<

woos

eaonmmuna
huamsoucH
suaaanmuame<

sua>auu<

mamom

meadow muonfiuou< .m.H.H.z mo huaocomoumuom EmuH one muamoowoaom madam

o mHan

39

in order to assess the degree of interrelationship of these scales (see
Table 7). Correlations ranged from .61 to -.33. With the exception of
the Adaptability-Approach and Adaptability-Mood and Mood-Approach rela-
tionship, the intercorrelations fall within the :_.35 range. This may
be interpreted as indicating low to moderate degrees of relationShip
between most of the attribute scales and suggests that the scales do mea-
sure substantially different domains and are not redundant. Yet, they
are clearly not orthogonal. This follows the assumption of Thomas, Chess
and Birch (1968) that it is "highly probable that they (the nine temper—
amental characteristics) are correlated with each other to some degree"
(p. 62). The three higher correlations indicate some overlapping of
content. Whether this is due to invalid items or reflects a true rela-
tionship between the hypothesized attributes requires further research.

The persistence scale did not approach the .70 coefficient alpha
criterion. In order to assess why this level of internal consistency
could not be reached, a factor analysis employing a varimax rotation
was conducted (see Appendix 4 for a complete listing of the items and
their loadings on the factors, as well as coefficient alpha and the
range of item-total correlations for each factor). This yielded six
factors with eigenvalues greater than 1.0 (see Table 8).

The inability to attain an acceptable level of internal consist-
ency as well as the presence of six factors clearly indicates a lack
of homogeneity within the scale. This could be accounted for in two
ways: 1) Persistence is a unitary construct that was not properly mea-
sured by the M.I.T.S. persistence scale; or 2) The attribute of per-
sistence-attention span as prOposed by Thomas and his colleagues is

multi-dimensional. The first explanation is difficult to test out. The

40

 

huwoaanuhnm

memo.

moao.
aaoaao.
«ammoc.
«awmmu.
«seamm.I

mNNH.I

 

 

 

 

 

maao.u mmHo.- «wees. maeo.u «mama. oewo.
same. memo. meH.I osmo. seesaw.

«aeomm. amms. .«eemm.n easea.u

ammo. «emwh~.- NmHo.-

««esm~.- «ammma.u
«ammmm.

Nuasapauomuuman mmmmmmma. woo: eaonwmuaa suamcmuam sua>aao<

suaaaamuaaea

Ho.u.mra
mo.uvmr

aufioaanuaam
suaaaaauumuuman
sumouoo<

woos

oaosmmusa
huwmsousH
soa>auo<

suaaanmuamea

Aooatmv mamom unassumnaoe odomaH cmwwnuwz can no moamom musowuuu< unwam mo msowumaouuooumucH

a «Home

41

was

m

n

mecca new: suaaanaxmam

ouamuo>omuma
mmmao>wumumoooo
aha Ou mosmoaou
mmosHkum

muuonno mo wcwxoom

 

 

uouomhhwmwswmoa Hosea Houomm

maouH mo

nonasz

.ooamm_mmamufi om co momma mammammm mouomm

.uouomm some now snowumamuuoo Hmuoulamufi
mamas udpwuawmmou mam Houomm mHmom mocmumwmnmm game so mwofiomoa Macao mam mamua pom m xfiocmmn< comm

 

 

m.wn q.w

m.o~ H.oa

m.oo o.HH

«.mc q.HH

H.wm m.mH

q.- «.mm
ammucoouom m>aumH5550 mommauw> mo soauuomoum

moamom swam soauamuu< mam monoumfimnom mo mfimhamc< Houomm

m «Haas

 

.ouoz

uouomm

42

persistence scale had face validity. Given its large number of items
(30) it should have been possible to extract the true persistence items
and construct an internally consistent scale.

It should be noted that other investigators have also encountered
difficulties with this scale. Their data suggest that our second explan-
ation is more appropriate. Thomas et al. (1968) reported some diffi-
culty with the temporal stability of their persistence scale. Bailey
(1977) also reported some problems in assessing persistence with the
Carey scale. Wilson and Lewis (1977) concluded that it was not a viable
concept due to poor definition and lack of stability over time.

The second possibility is that persistence, as an attribute, is com-
posed of more than one temperamental component. This explanation may
account for the poor temporal stability that has been reported.

Rothbart, Furby, Kelly, and Hamilton (1977) endorse this idea, pr0pos-
ing two persistence attributes: undisturbed persistence which refers

to visual or behavioral interaction with an object where no sudden

change in stimulation occurs, and obstacle persistence which describes

an infant's seeking of an object after it has been prohibited or moved
out of the way. The latter construct seems to describe the items in
Factor 1 of the M.I.T.S. persistence scale, "seeking of objects". Undis-
turbed persistence may correspond to Factor 2 ("stillness"). The remain-
ing four factors of the persistence scale may be situational or behavior~
specific variables, or error factors. Statistical criteria for factor
analytic solutions suggests the latter explanation is most appropriate.
That is, employing Cattell's (1966) scree test, only factors 1 and 2
would be considered "non-trivial factors" as the next four factors, when

plotted, compose a relatively straight scree line.

43

Insofar as there is any meaningful variance, the following explan-
ations make some sense: Examination of the items that make up factors 3-
6 suggests some additional similarities with two other attributes pro-
posed by Rothbart et al.: soothability and anger/frustration. Factor 3
("tendency to cry") can be described as persistence or duration of
affect. This may correspond to Rothbart's soothability which also seems
to describe the length and modifiability of an infant's feelings. The
last three factors while each seeming to tap specific temperamental
aspects ("cooperativeness," "perseverance," and "flexibiltty with foods"
may in fact interrelate along the dimension of anger/frustration. Many
of the items in Factors 4, 5, and 6 deal with reactions to frustrations
in feeding or motor activity. Therefore, these factors may group togeth-
er as a second order factor tapping anger/frustration.

The weight of the evidence questions the viability of the persist-
ence-attention span attribute as suggested by Thomas et al. It is
proposed that this construct might be better explained and measured by
breaking it up into its component parts. While Thomas et al. describe
a single construct as tapping selectivity of attention, frustration tol-
erance, and persistence, Rothbart et al. suggest two separate attributes.
It may also be possible to explain this behavioral style employing com-
binations of the other eight attributes, i.e., the interaction of act-
ivity level and distractibility may describe selectivity of attention
and persistence while mood might account for frustration tolerance.

The original M.I.T.S. persistence scale is not an internally con-
sistent measure and thus will be excluded from the revised form of the
M.I.T.S. However, it is clear that further investigation is necessary

to explore the behavioral characteristics commonly termed persistence.

44

In summary, drawing from the 243 item pool of the M.I.T.S., a
revised instrument, (M.I.T.S.-2) containing 136 items tapping eight
temperamental attributes (17 items each) was constructed. Each attri-
bute scale achieved at least a minimum level of internal consistency
(.705 coefficient alpha) and was judged by the author to contain a
broad range of infant behaviors and situations. It should be noted
that further samplings involving analysis of internal consistency is
necessary to establish the accuracy of the above reliability levels.

Temporal Stability

 

Test-retest correlations for the "two-week" and "eight-week"
intervals using the revised 136 item measure are listed in Table 9.

The two-week retest correlations range from r= .62 to .89 indicating
a high degree of short-term stability in the attribute scores. The
eight-week retest yielded more moderate relationships with the corre-
lations ranging from .41 to .64. This indicates that there is some
change in scale scores over an eight week period; yet a basic consist-
ency in subjects' scores does appear to exist.

The levels of the above retest correlations approximate those
obtained by Rothbart, Furby, Kelly, and Hamilton (1977), Birns, Barten
and Bridger (1969), and Thomas, Chess, Birch, Hertzig, and Korn (1963).
The loss in some predictability over an eight week period may in fact be
due to instability in the actual behavior characteristics. However, it
seems likely that the changes in scores on the M.I.T.S. are affected by
other factors such as the large number of items which may have induced
fatigue of inattention in the respondents. Also, some of the items may
be too age-specific to gain repeated endorsement during the period of

rapid behavior change characterized by infancy.

45

.uouumn so No.mm um unmofiwwcwfim mum m:0fiumamuuoo HH< "muoz

one.
now.
own.
moo.
wmq.
mmq.
wan.

«co.

 

Ammmmv nsoum xom3lu£wam

mac.
mmo.
mwo.
own.
own.
Cum.
Hum.

oww.

 

Anmtmv dsouw xmoanae

eooz
shamanauumuumao
muwmsoucH
suaaanuuame<
suaoaaausnm
nomouoa<
sua>auu<
aaoammuna

oamom

moamom ousofiuuu< .m.H.H.z mo mcoaumamuuoo ummquIumoH somBIunmfim can 039

a magma

46

Validity

Inter-rater comparisons were employed as crude indicators of con-
struct and scale validity. In Chapter I, it was stated that temperament
"refers to a set of underlying formal characteristics which influences
or determine behavioral style" (p. 3). This definition implies a con-
sistency in style across situations and with different people. Thus
it would be predicted that there would be a significant relationship
between mother's ratings and that of father or an alternate caretaker.
While a low or negative correlation would leave the issue of concept
validity unclear (the scale may be measuring characteristics less stable
than temperament), a significant positive relationship would suggest
a cross-situational or inter-observer consistency in the rated charac-
teristics.

Before an inter-observer comparison can be done, the informant's
knowledgeability about the infant must be assessed. Throughout this
study, it has been assumed that the mother is quite well informed
about her child. If the mother is the primary caretaker, it is pos-
sible that the father may have less contact with and information about
his child. A babysitter or alternate caretaker is likely to have only
intermittent contact with the infant. Although caretaker knowledge
could not be assessed, three hours of infant care per week was set as
a minimum amount of contact necessary for informed caretaker rating.
Fathers' and mothers' time with their infants was assessed and is shown
in Table 10- Mothers clearly spend more time with their infants than
fathers both during the week and on the weekend. However, it does
appear that on the average the fathers do spend a significant amount of

time with their infants on weekdays and weekends (X=3.39 and 4.88,

47

 

 

 

 

mm.mJM mw.qmm. mn.mJM am.mJM
IMJMMI IMJMMI IWWMWI w.mH whoa so o.o
q.e m.~ N.m N.m m.m
w.m H.- a.w o.Hm o.m ou m.m
o. N.m o. H.qm o.m ou m.H
o w.m o o.~a o.H 0» n.
m.H q.q o.H ~.m o
Auaoouomv Ausmoummv ham mom muoom
Honuoz umnumm posuoz nonumm
ammo vamxmoz mxmoxooz

mucmmsH Macao noes moo mom musom whoa no 0 co 0 wsfiocoam muosuoz was muonumm mo mwmucoouom

OH mHan

48

respectively . Therefore, most fathers as well as mothers are likely
to be well informed about their child's behavior and temperamental
style.

Table 11 contains the correlation between maternal ratings and
those of father and caretaker. Many of the scales show significant
correlations in both the mother-father and the mother-caretaker compar-
isons. The insignificance of some of the correlations as well as the
low relationship of Activity and Approach in the mother-father compar-
ison are likely to have been influenced by the low number of observ-
ations. However, the lower correlations may also be due to character-
istics of the observer and his/her opportunities to observe the infant.
The low relationship in the mother-father and mother-caretaker compar-
ison on Activity may reflect demand characteristics of these two
"alternate caretakers" as well as a skewed sampling of the child's
behavior. Having a more limited amount of time with the child and not
being involved in the full range of infant-care duties would make fath-
er and caretaker ratings less reliable with respect to mother ratings.
Similar factors may be operating in the mother-father correlations
for Approach and mother-caretaker comparison of the Rhythmicity scale.
An alternate caretaker (babysitter or day care teacher) will not be
exposed to the full range of infant behaviors centering around feeding,
toileting, and sleeping. Similarly, fathers are less likely to observe
their infants during initial exposures to new stimuli; bathes, doctors,
new foods, etc. are often presented to a child when the father is at
work. However, the above explanations remain speculative. Explorations
of these low relationships requires more detailed information about

father and caretaker activities with the infants, work schedules, mother

49

30. v N33.
3. v Na:
no.uhmra
ca.vmmr

.oo.H cmnu Houmouw on ou mnofiumaouuou omuoouuoo now asoaam
ousoooouo coaumasouuo pom soauuouuoo onu mo ousumm was asap mouoa on masons uH
.uoumsfiaocoo ca mcoaumaouuoo amouou zoosIunwao moan: coaumsaouum How nouuouuoum

 

 

 

Aanm. v was. Ana». e «.mem. suaoaaausam
Aoae. v mmm. Ashe. V egaam. suaaaaeuumuuman
Ames.av «aoam. Anna. v moo. nuaouaaa
Ammo.av «owe. AeH~.HV «aasmmm. moo:
Anne. V «awe. AqH~.Hv «aaamfin. oHosmounH
Aomm. V «aman. Anus. v Hum. augmnousH
32. V SH. 33 . V So. 53.3%

Ana.av «was. «Amen. e «seem. suaaapuuaan<

Asauzv uasauuumuuuoauoz mamuzv penummuumnuoz mannauuu<

.m.H.H.z onu mo moaoom ousoauuud unwam now osoaumaouuou uo>nomn0IuoucH

HH mHan

50

and father parenting roles, etc.

As described above, the mean time spent by fathers with their
infants is lower than that spent by mothers. In order to investigate
whether the time spent by fathers had an effect upon the agreement of
father and mother scores on the M.I.T.S., comparison of the means for
each attribute scale and correlations of difference scores with father's
time spent with the infant were conducted. Table 12 contains these data.
It can be seen that the means of most of the attribute scales are not
significantly different for the two sets of raters. There is also lit-
tle relationship between time spent by the father with his infant and
the level of agreement between fathers and mothers. Only the intensity
scale has a significant correlation (r=-.36). This suggests that as a
father spends more time with his child, his rating of the strength of
the infant's response goes down. This may be a result of sensory adapta-
tion. That is, fathers who spend more time with their infants become
used to the intensity of his/her reactions. It is also possible that as
the time spent by father increases, the infant's strength of response
decreases. This might be due to greater familiarization and comfort
with the father. Finally, it is possible that this correlation is spur-
ious and non-replicable. Evaluation of the effects of time upon both
mother and father's ratings should be conducted on future samples.

Temperament theory postulates cross-situational and inter-observer
behavioral consistency. The above results are a beginning step in sug-
gesting that this basic consistency exists, at least for some of the
attribute scales. However, more in-depth assessment of inter-observer

similarities and differences is clearly necessary.

51

 

ca.I am.al
ao. m¢.a
oa.I am.I
mo.I mm.
ao. «wN.N
«om.I «m.~I
ca.I cq.al
ca. hm.a
oaau suaa museum MMMMMJH

mucouomwao mo soaumaouuou

 

New. on».
mac. nmo.
mmw. amm.
nan. awn.
«mo. qao.
Nam. owe.
mm“. own.
mum. mam.
umnuoz smooch

ouoom one:

no.”.mr

muaofianuhgm
suaaaaauumuumaa
nomouao<

coo:

eaoammuas
muamsousa
5:33.
suaaaamuamea

oamom

unmwca mag zuaB Hannah an unmow oaaa nuas mmuoom monouomman oawom mo soaumamuuoo

was some: oamom .m.H.a.z mo canoe ca mwsaumm nonuoz can summon mo msomaumoaoo

Na canoe

52

Further Considerations

 

The purpose of this study has been to construct an instrument for
the measurement of infant temperament. Initial data indicate that the
eight attribute scales of the M.I.T.S. are internally consistent. Their
short-term temporal stability (two and eight-week) is relatively high.
This suggests that the instrument does measure certain behavioral char-
acteristics well.4 However, this is based upon one sample which served
the dual role of: 1) providing item inter-correlations for test con-
struction; and 2) establishing levels of internal consistency in the
final versions of the scales. The necessary "playing around" with the
test items to achieve acceptable levels of test length and internal con—
sistency involves a certain amount of taking advantage of chance
(Nunnally, 1967). A re-analysis utilizing a new sample of respondents
is necessary to clearly establish the extent to which the initial item
analysis capitalized on sampling errors. It remains to be seen whether
this will demonstrate small sampling error and continual high levels of
scale homogeneity.

A second issue in scale reliability is that of temporal stability..
The study suggests that attribute scores are relatively stable over
short periods of time. However, the subsample sizes were small. Fur—
ther test-retest analyses would allow greater confidence in the consist-
ency of the M.I.T.S. scores. In fact, temporal stability is also an

issue of construct validity (see page 4). MOre extensive evaluation of

4It is assumed that these characteristics parallel the temperamental

attributes proposed by Thomas, Chess, Birch, Hentzig and Korn (1963)
although the question of scale validity has been formally pursued.

53

the stability of attribute scale scores over greater lengths of time
would yield information regarding the permanence or consistency of
behavioral styles. Theory would suggest at least moderate correlations
of scale scores for a retest period of one year or more; although devel-
opmental change would also appropriately be expected. Failure to obtain
such a relationship would cast doubt upon the utility and validity of
the temperament construct.

Finally, the issue of validity needs to be further explored both
through the use of larger samples of mother-father and mother-caretaker
comparisons, as well as through the use of known-groups, convergent,
and divergent validation techniques. The present data was obtained from
samples too small to allow much weight to be given to the resulting cor-
relations. Viewed as trends, the inter-observer comparisons indicate
some consistency of ratings by different well-informed raters. Larger
samples are necessary to establish whether or not all eight scales
share this cross-situational and inter-observer stability of behavioral
style.

In summary, the M.I.T.S. appears to hold promise as an instrument
for measuring eight temperamental attributes. However, at this time
the scale must be viewed as an experimental instrument whose character-
istics need to be further investigated. Scale reliability needs to be

confirmed while construct and scale validity remains to be examined.

LIST OF REFERENCES

 

III - It!" 1 II I

I‘ll ill.

. I. I'lfvld it! IIIlAYILrIl l..1l...lll, . I .. III. I) r.

. lultla..‘ll||‘l

LIST OF REFERENCES

Allport, G.W. Pattern and growth in personality. New York: Holt,
Rinehart and Winston, 1961.

Bailey, M.M. An observational study of mothers' commands and their
young children's responses. Unpublished doctoral dissertation,
Michigan State University, 1977.

Birns, B., Barten, S., & Bridger, W.B. Individual differences in temp-
eramental characteristics of infants. Transactions of the New York
Academy of Science, 1969, 31(8), 1071-1082.

 

 

Bronson, W.C. Stable patterns of behavior: The significance of endur-
ing orientations for personality develOpment. In Hill, J.P. (Ed.),
Minnesota Symposia on Child Psychology (Vol. II). Minneapolis:
University of Minnesota Press, 1969.

Bronson, W.C. Central orientations: A study of behavior organization
from childhood to adolescence. In Jones, M.C., Bayley, N.,
MacFarlane, J.W., Honzik, M.D. (Eds.), The Course of Human Develop-
ment. Waltham: Xerox College Publications, 1971.

 

Buss, A.H., Plomin, R., & Willerman, L. The inheritance of temperament.
Journal of Personality, 1973, 41(4), 513-524.

 

Buss, A.H., & Plomin, R. A temperament theory ofgpersonality develop-
ment. New York: John Wiley & Sons, 1975.

Carey, W.B. A simplified method for measuring infant temperament.
Journal of Pediatrics, 1970, 71, 188-194.

Carey, W.B. Measuring infant temperament. Journal of Pediatrics, 1972,
.§l(2), 414.

 

Carey, W.B. Clinical applications of infant temperament measurements.
Journal of Pediatrics, 1972b, 81(4), 823-828.

 

Carey, W.B. Night waking and temperament in infancy. Journal of Pedi-
atrics, 1974, 84(5), 756-758.

 

Carey, W.B. Personal communication. December, 1976.

Carey, W.B., Lipton, W.L., & Myers, R.A. Temperament in adopted and
foster babies. Child Welfare, 1974, 23(6), 352-359.

 

54

55

Cattell, R.B. The scree test for the number of factors. Multivariate
Behavioral Research, 1966,_l(l), 245-276.

 

Diamond, S. Personality and temperament. New York: Harper and
Brothers, 1957.

Erikson, E.H. Childhood and society. New York: W.W. Norton & Company,
1950.

Feiring, C., & Taylor, J. The influence of the infant and secondary
parent on maternal behavior: Toward a social systems view of infant
attachment. Unpublished manuscript. University of Pittsburgh,
1976.

Freud, A. The ego and the mechanisms of defense. London: Hogarth
Press Ltd., 1948.

Freud, S. A general introduction to psycho-analysis. New York: Simon
and Schuster, 1969.

Gessell, A., & Ames, L.B. Early evidences of individuality in the
human infant. Journal of Genetic Psychology, 1937, 41, 335.

Graham, P., Rutter, M., & George, S. Temperamental characteristics
on predictors of behavior disorders in children. American Journal
of Orthopsychiatry, 1973, 43, 328-339.

 

Heymans, G. Ueber einige psychischen korrelationen, zeitschr. f. argew.
Psychology, 1908, 1, 313-381.

Hollingshead, A.B. Two factor index of social position. New Haven:
S.N., 1957.

 

Levandovskii, N.G. On the structure of personality. Voprosyflpsikho-

 

McDevitt, S. & Fox, M. Reliability and construct validity of the Carey
Infant Temperament Scale: A reanalysis. Unpublished manuscript.
Temple University, 1976.

Nunnally, J.C. Psychometric theory. New York: McGraw-Hill, 1967.

Pederson, F.A., Anderson, B.J., & Cain, Jr., R.L. A methodology for
assessing parental perception of infant temperament. Paper pre-
sented at the Fourth Biennial Southeastern Conference on Human
Deve10pment, April, 1976.

Rothbart, M.K., Furby, L., Kelly, S.R., & Hamilton, J.S. Development
of a caretaker Report Temperament Scale for use with 3-, 6-, 9-, and
12-month old infants. Paper presented at Society for Research in
Child Development meetings, March, 1977.

56

Scarr, S. & Salapatek, P. Patterns of fear development during infancy.
Merrill-Palmer Quarterly of Behavior and Development, 1970, 16,
53-90.

 

Scholom, A.H. The relationship of infant and parent temperament to the
prediction of child adjustment. Unpublished doctoral dissertation,
Michigan State University, 1975.

Sheldon, W. The varieties of temperament: A psychology of constitution-
al differences. New York: Harper, 1942. ‘

 

Shirley, M. The first two years: A study of twenty—five babies (Vol.
III: Personality manifestations). Minneapolis: University of
Minnesota Press, 1933.

 

Sullivan, H.S. The interpersonal theory of psychiatry. New York:
W.W. Norton & Co., 1953. '

Thomas, A., Chess, 8., Birch, H. & Hertzig, M. A longitudinal study of
primary reaction patterns in children. Compgehensive Psychiatry,
1960, 1, 103-112. -

 

Thomas, A., Chess, 8., Birch, B.G., Hertzig, M.E., & Korn, S. Behavior
individuality in early childhood. New York: New York University
Press, 1963.

 

Thomas, A., Chess, 8., & Birch, H.G. Temperament and behavior disorders
in children. New York: New York University Press, 1968.

 

 

U.S. Bureau of the Census. Census of Population: 1970 Detailed Char-
acteristics, Final Report Pc (l)-D24 Michigan.’ U.S. Government
Printing Office, Washington, D.C., 1972.

Wilson, C.D. & Lewis, M. Temperament: A developmental study in stabil-
ity and change during the first four years of life. Unpublished
manuscript. Educational Testing Service, 1977.

APPENDICES

APPENDIX A

APPENDIX A

Carey Infant Temperament Survey

Age of Child

 

Number

 

The purpose of the enclosed questionnaire is to determine the gen-
eral pattern of your baby's reaction to his or her environment by _
getting specific information about many areas of functioning. You will
also be asked some questions about that environment and about your gen-
eral impressions of the baby.

The temperament questionnaire itself consists of 70 statements
about the baby, each with three choices. Please circle the letter "a",
"b", or "c" before the choice that best describes the baby. There are
no good and bad or right and wrong answers, only descriptions of what
your baby is like.

57

Survey of Temperamental Characteristics

 

Generally goes to sleep at about same time (within half an
hour) night and naps.

Partly the same times, partly not.

No regular pattern at all. Times vary 1-2 hours or more.

Generally wakes up at about same time, night and naps.
Partly the same times, partly not.
No regular pattern at all. Times vary 1-2 hours or more.

Generally happy (smiling, etc.) on waking up and going to
sleep.

Variable mood at these times.

Generally fussy on waking up and going to sleep.

Moves about crib much (such as from one end to other) during
sleep.

Moves a little (a few inches).

Lies fairly still. Usually in same position when awakens.

With change in time, place or state of health:

Sleep

69
(b)
(e)
2. (a)
(b)
(e)
3. (a)
(b)
(C)
4. (a)
(b)
(e)

5.
(a)
(b)
(C)

Feeding

6. (a)
(b)
(e)
7. (a)
(b)
(e)
8. (a)
(b)
(e)
9. (a)
(b)
(e)

Adjusts easily and sleeps fairly well within 1-2 days.
Variable pattern.

Bothered considerably. Takes at least 3 days to readjust
sleeping routine.

Generally takes milk at about same time. Not over 1 hour
variation. '

Sometimes same, sometimes different times.

Hungry times quite unpredictable.

Generally takes about same amount of milk, not over 2 oz. dif-
ference.

Sometimes same, sometimes different times.

Amounts taken quite unpredictable.

Easily distracted from milk feedings by noises, changes in
place or routine.

Sometimes distracted, sometimes not.

Usually goes right on sucking in spite of distractions.

Easily adjusts to parents' efforts to change feeding schedule
within 1-2 tries.

Slowly (after several tries) or variable.

Adjusts not at all to such changes after several tries.

58

10.

11.

12.

13.
14.

15.

1‘35.-

17.

18.

19.

20.

(a)

(b)
(C)

(a)

(b)
(C)

(a)

(b)
(C)

(a)
(b)
(C)
(a)
(b)
(C)
(a)
(b)
(C)
(a)

(b)
(C)

(a)
(b)
(C)
(a)

(b)
(C)

(a)

(b)
(C)

(a)

(b)
(C)

59

If hungry and wants milk, will keeping refusing substitutes
(solids, water, pacifier) for many minutes.

Intermediate or variable.

Gives up within a few minutes and takes what is offerred.

With interruptions of milk or solid feedings, as for burping,
is generally happy, smiles.

Variable response.

Generally cries with these interruptions.

Always notices (and reacts to) change in temperature or type
of milk or substitution of juice or water.

Variable. -

Rarely seems to notice (and react to) such changes.

Suck generally vigorous.
Intermediate.
Suck generally mild and intermittent.

Activity during feedings--constant squirming, kicking, etc.
Some motion: Intermediate.
Lies quietly throughout.

Always cries loudly when hungry.
Cries somewhat but only occasionally hard or for many minutes.
Usually just whimpers when hungry, but doesn't cry loudly.

Hunger cry usually stopped for at least a minute by picking up,
pacifier, putting on bib, etc.

Sometimes can be distracted when hungry.

Nothing stops hunger cry.

After feeding baby smiles and laughs.
Content but not usually happy (smiles, etc.) or fussy.
Fussy and wants to be left alone.

When full, clamps mouth closed, spits out food or milk, bats
at spoon, etc.

Variable.

Just turns head away or lets food drool out of mouth.

Initial reaction to new foods (solids, juices, vitamins)
acceptance. Swallows them promptly without fussing.
Variable response.

Usually rejects new foods. Makes face, spits out, etc.

Initial reaction to new foods pleasant (smiles, etc.), whether
accepts or not.

Variable or intermediate.

Response unpleasant (cries, etc.), whether accepts or not.

21.

22.

23.

24.

25.

(a)
(b)
(C)

(a)
(b)
(C)
(a)

(b)
(C)

(a)

(b)
(C)

(a)

(b)
(C)

60

This response is dramatic whether accepting (smacks lips,
laughs, squeals) or not (cries).

Variable.

This response mild whether accepting or not. Just smiles,
makes face or nothing.

After several feedings of any new food, accepts it.
Accepts some, not others.
Continues to reject most new foods after several tries.

With changes in amounts, kinds, taming of solids, does not
seem to mind.

Variable response. Sometimes accepts, sometimes not.

Does not accept these changes readily.

Easily notices and reacts to differences in taste and consist-
ency.

Variable.

Seems seldom to notice or react to these differences.

If does not get type of solid food desired, keeps crying till
gets it.

Variable.

May fuss briefly but soon gives up and takes what offerred.

Soiling and Wetting

 

26.

27.

28.

29.

30.

31.

(a)
(b)
(C)

(a)

(b)
(C)

(a)
(b)
(C)

(a)
(b)
(C)

(a)
(b)
(C)

(a)
(b)
(C)

When having bowel movement, generally cries.

Sometimes cries.

Rarely cries though may get red in face. Generally happy
(smiles, etc.).

Bowel movements generally at same time of day (usually within
1 hour of same time).

Sometimes at same time, sometimes not.

No real pattern. Usually not same time.

Generally indicated somehow that is soiled with b.m.
Sometimes indicates.
Seldom or never indicates.

Usually fusses when diaper soiled with b.m.
Sometimes fusses.
Usually does not fuss.

Generally indicates somehow that is wet (no b.m.).
Sometimes indicates.
Seldom or never indicates.

Usually fusses when diaper wet (no b.m.).
Sometimes fusses.
Usually does not fuss.

32.

33.

Diapering

61

(a) When fussing about diaper, does so loudly. A real cry.
(b) Variable.

(C)
(a)

(b)
(C)

Usually just a little whimpering.

If fussing about diaper, can easily be distracted for at
least a few minutes by being picked up, etc.

Variable.

Nothing distracts baby from fussing.

 

34.

35.

36.

(a)
(b)
(C)

(a)

(b)
(C)

(a)
(b)
(C)

Bathing

37.

38.

39.

40.

41.

42.

(a)
(b)
(C)

(a)
(b)
(C)
(a)
(b)
(C)
(a)
(b)
(a)
(b)
(C)
(a)

(b)
(C)

and dressing

Squirms and kicks such at these times,
Moves some.
Generally lies still during these procedures.

Generally pleasant (smiles, etc.) during diapering and dres-
sing.

Varied.

Generally fussy during these times.

These feelings usually intense: vigorous laughing or crying.
Varied.
Mildly expressed usually. Little smiling or fussing.

Usual reaction to bath--smiles or laughs.
Variable or neutral.
Usually cries or fusses.

Like or dislike of bath is intense. Excited.
Variable or intermediate.
Like or dislike is mild. Not very excited.

Kicks, splashes and wiggles throughout.
Intermediate-~moves moderate amount.
Lies quietly or moVes little.

Reaction to very first tub (or basin) bath. Seemed to accept
it right away.
At first protested against bath.

If protested at first, accepted it after 2 or 3 times.
Sometimes accepted, sometimes not.
Continued to object even after two weeks.

If bath by different person or in different place, readily
accepts change first or second time.

May or may not accept.

Objects consistently to such changes.

62

Procedures - nail cutting, hair brushing, washing face and hair,

43. (a)
(b)
(C)

44. (a)
(b)
(C)
45. (a)

(b)
(C)

46. (a)

(b)
(C)

Visits to

medicines

Initial reaction to any new procedure--generally acceptance.
Variable.
Generally objects; fusses or cries.

If initial reaction to any new procedure--generally acceptance.
Variable.
Continues to object even after several times.

Generally pleasant during procedures once established--smiles,
etc.

Neutral or variable.

Continues to object even after several times.

If fussy with procedures, easily distracted by game, toy, sing-
ing, etc.--and stops fussing.

Variable response to distractions.

Not distracted. Goes on fussing.

doctor

 

47. (a)
(b)
(C)

48. (a)
(b)
(C)
49. (a)

(b)
(C)

With physical exam, when well, generally friendly and smiles.
Both smiles and fusses: variable.
Fusses most of the time.

With shots cries loudly for several minutes or more.
Variable.
Cry over in less than a minute.

When crying from shot, easily distracted by milk, pacifier,
etc.

Sometimes distracted, sometimes not.

Goes right on crying no matter what is done.

Response to illness

 

50. (a)
(b)
(C)
Sensogy -
51. (a)
(b)
(C)
52. (a)
(b)
(C)

With any kind of illness much crying and fussing.

Variable.

Not much crying with illnesses. Just whimpering sometimes.
Generally his usual self.

reactions to sounds, light, touch

Reacts little or not at all to unusual loud sound or bright
light.

Intermediate or variable.

Reacts to almost any change in sound or light.

This reaction to light or sound is intense--start1es or cries
loudly.

Intermediate--sometimes does, sometimes not.

Mild reaction--1ittle or no crying.

53.

54.

(a)

(b)
(C)

(a)

(b)

(C)

63

On repeated exposure to these same lights or sounds, does not
react to much any more.

Variable.

No change from initial negative reaction.

If already crying about something else, light or sound makes
crying stop briefly at least.

Variable response.

Makes no difference.

Response to people

 

55.

56.

57.

58.

59.

(a)

(b)
(C)

(a)

(b)
(C)

(a)

(b)
(C)

(a)

(b)
(C)

(a)

(b)
(C)

Definitely notices and reacts to differences in people: age,
sex, glasses, hats, other physical differences.

Variable reaction to differences.

Similar reactions to most people unless strangers.

Initial reaction to approach by strangers positive, friendly
(smiles, etc.).

Variable reaction.

Initial rejection or withdrawal.

This initial reaction to strangers is intense: crying or
laughing.

Variable.

Mild--frown or smile.

General reaction to familiar peOple is friendly--smiles,
laughs.

Variable reaction.

Generally glum or unfriendly. Little smiling.

This reaction to familiar people is intense--crying or laugh-
ing. .

Variable.

Mild--frown or smile.

Reaction to new places and situations

 

60.

61.

(a)
(b)
(C)
(a)

(b)
(C)

Initial reaction acceptance--tolerates or enjoys them within
a few minutes.

Variable.

Initial reaction rejection--does not tolerate or enjoy them
within a few minutes.

After continued exposure (several minutes) accepts these
changes easily.

Variable.

Even after continued exposure, accepts changes poorly.

63.

64.

65.

66.

67.

68.

69.

(a)
(b)

(C)

(a)
(b)
(C)

(a)

(b)
(C)

(a)

(b)
(C)

(a)

(b)
(C)

(a)
(b)
(C)

(a)
(b)
(C)

(a)
(b)
(C)

(a)
(b)
(C)

64

In crib or play pen can amuse self for half hour or more look-
ing at mobile, hands, etc.

Amuses self for variable length of time.

Indicates need for attention or new occupation after several
minutes.

Takes new toy right away and plays with it.
Variable.
Rejects new toy when first presented.

If rejects at first, after short while (several minutes)
accepts new toy.

Variable.

Adjusts slowly to new toy.

Play activity involves much movement--kicking, waving,arms,
etc. Much exploring.

Intermediate.

Generally lies quietly while playing. Explores little.

If reaching for toy out of reach, keeps trying at it for 2
minutes or more.

Variable.

Stops trying in less than 1/2 minute.

When given a toy, plays with it for many minutes.
Variable.
Plays with one toy for only short time (only 1-2 minutes).

When playing with one toy, easily distracted by another.
Variable.
Not easily distracted by another toy.

Play usually accompanied by laughing, smiling, etc.
Variable or intermediate.
Generally fussy during play.

Play is intense: much activity, vocalization or laughing.
Variable or intermediate.
Plays quietly and calmly.

65

Table 13

Scoring Key for Carey Infant Temperament Scale

Scale Items
Activity 4, 13, 14, 34, 39
Rhythmicity 1, 2, 6, 7, 27
Approach 19, 401, 43, 56, 60, 63
Adatability 5, 9, 22, 23, 35, 41, 42, 44, 47, 53, 61, 64
* * * * *
Threshold 12 , 24 , 28 . 30 , 42, 51, 55
*
Intensity 15, 18, 21, 26 , 32, 36, 38, 48, 52, 57, 59, 7o
*
Mood 3, 11, 17, 20, 26, 292, 312, 35, 37, 45, 47, 50 ,
56, 58, 60, 69
Distractibility 8, 16, 33, 46, 49, 54, 68

Persistence 10, 25, 62, 66, 67

Scoring Instructions:

00"!“
II II II
(AMI-i

(1) Do not score "b"

(2) Do not score c

* Reverse "a" and "c

II,
I

c=l a=3

APPENDIX B

APPENDIX B

GENERAL INSTRUCTIONS
Dear Parents:

Thank you for participating in our study Of infant behavior. As
you know, we are interested in the way infants 3 to 12 months Of age
behave. Enclosed you will find a questionnaire packet designed to
gather information from you, the person most knowledgeable about your
infant's behavior.

The questionnaire should take you no more than 1 hour to complete.
In fact, most individuals require 45 minutes or less. It is extremely
important that you do the questionnaire in one sitting. In other words,
wait until you have 45 minutes Of time where you are unlikely to have
many distractions or responsibilities that would compete with complet-
ing the forms. Many parents find that naptime or after the children are
in bed to be convenient times for this. Please do the questionnaire
alone, without consulting others on the answers to any of the items.

You should have four (4) separate forms and an addressed, stamped
envelope in the packet sent to you. These forms are: Departmental
Research Consent Form; Home Information Sheet; M.I.T.S. questionnaire;
and Supplementary Information Form. Please complete these forms in the
following order and in exact accordance with the instructions:

GO TO THE DEPARTMENTAL RESEARCH CONSENT FORM

1. Completely read the consent form. Please read and sign
your name and record the date at the bottom Of this page.
All identifying information will be removed from the
forms and the results will be held completely confiden-
tial. By doing this work now, you are under no obligation
to participate in the follow-up aspects Of the study.

2 NOW 80 TO THE HOME INFORMATION SHEET (next page)

2. On the Home Information Sheet, record your infant's first
name, your relationship to your infant (mother or father),
your child's date of birth, and his or her sex.

NOW TURN TO THE QUESTIONNAIRE (M.I.T.S.)

3. Put the above two forms in the stamped addressed return
envelope and set aside. Now, carefully read the
instructions on the front of the M.I.T.S. questionnaire.
When you have finished reading the instructions, begin
the questionnaire. Read each item and put your answer
(T or F) to the left Of the item number. When you fin-
ish the questionnaire, put it in the return envelope.

66

67

GENERAL INSTRUCTIONS letter

Page 2

GO TO THE M.I.T.S. SUPPLEMENTARY INFORMATION FORM

Finally, complete the M.I.T.S. Supplementary Information
Form. When you are done put the form in the return packet,
and mail it back tO us. The envelope is pre-addressed
and-stamped so you need only place it in a local mailbox.
If you have requested information about the study and/or
expressed interest in participating in a follow-up ques-
tionnaire, we will contact you within the next 3 months.
However, if you have any questions, contact us at any time
at the address above, or by phone (Mr. Bonem) at 355-9564
(8 to 5) or evenings at 332-2486. Again, thanks for your
OOOperation.

Sincerely,

Howard Bonem Robert Zucker, Ph.D.

HB/Rz:sh

Michigan State University
Department Of Psychology

DEPARTMENTAL RESEARCH CONSENT FORM

I have freely consented to take part in a scientific study being
conducted by: Howard Bonem under the supervision Of:
Dr. Robert A. Zucker, Academic Title: Professor

 

 

 

The study has been explained to me and I understand the explan-
ation that has been given and what my participation will involve.

I understand that I am free to discontinue my participation in the
study at any time without penalty.

I understand that the results Of the study will be treated in strict
confidence and that I will remain anonymous. Within these restrict-
ions, results Of the study will be made available to me at my
request.

I understand that my participation in the study does not guarantee
any beneficial results to me.

I understand that, at my request, I can receive additional explan-
ation Of the study after my participation is completed.

Signed

 

Date

 

68

HOME INFORMATION SHEET

Please provide us with the information requested below.

First name Of infant

 

Your relationship to infant: Mother Father (circle)

Infant's date of birth

 

Month - Day - Year

Sex Of infant: Male Female (circle)

69

M.I.T.S.

This questionnaire is designed to find out how your infant behaves
in a variety Of everyday situations. Read each item carefully and
decide whether it is true or mostly true, or false or mostly false for
your child. Mark your answer by putting a T (true) or F (false) beside
the item.

Example: 1A. Infant Often plays with his/her food.

If your infant Often plays with his or her food, then you would
mark your answer like this:

T_ 1A. Infant Often plays with his/her food.
Should your child not play with fOOd while eating, then you
would put an F beside the question like this:

F, 1A. Infant Often plays with his/her food.

Please read and try to answer all items. If a question is com-
pletely inappropriate, then you may omit it (be sure to skip the Space
on the questionnaire). If your child has outgrown an activity or
behavior mentioned in an item, answer the item according tO how he/
she used to act. That is, if your child drinks only from a cup,
answer items regarding breast and bottle-feeding from your memory of
his/her feeding habits.

We have worked very hard to make all items equally applicable to
infants Of both sexes and to those who have been breast or bottle-fed.
Please do the best you can in answering as accurately as possible.

All information on this questionnaire is confidential, and will
only be handled by the research staff, with no names attached. If you
have any questions, please contact one Of the examiners.

REMEMBER, ANSWER ALL QUESTIONS.

70

1G.

2G.

3G.

4G.

SG.

6G.

7G.

8G.

9G.

10G.

11G.

12G.

13G.

14G.

15G.

16G.

18G.

19G.

20G.

21G.

22G.

23G.

71

If child fusses from loss of a toy, he/she does not do so for very
long.

Child generally reacts to familiar persons in an unfriendly man-
ner--does not smile.

Baby does not persist in attempts at turning over, crawling, or
walking. Gives up easily.

The infant usually rejects a new toy when it is first presented.
Baby is irritable or cranky after sleep.

When baby is being fed, he grabs for the food when it comes near.
There is a great deal Of fussing and crying with any illness.
When waiting to be fed, baby is generally still.

After several feedings Of a new food, child accepts it.

Infant generally exhibits exploratory behavior on his/her own.

Responses to diapering and dressing are usually intense with much
laughing or crying.

When lights are turned on in the room, infant is usually not
awakened.

When lying in crib, child is usually looking all around.

Child does lots Of squirming or kicking while being diapered or
dressed.

Stops eating if hears noise such as yell, radio, etc.

Child wakes up most mornings at about the same time--within half
an hour.

During play the infant is usually very active and vocal.

When playing with one toy, the infant is easily distracted by
another.

Baby does not react to physical differences in familiar peOple.
Child kicks, splashes or giggles throughout bath.
Child usually fusses during diapering and dressing.

Infant responds to any small change in surrounding environment.

246.
256.

266.
27G.

28G.
29G.

306.

316.
326.
336.
346.
356.
366.
376.
386.
396.
406.

A 416.
426.

436.

72
There is no clearly evident pattern in the time for child's bow-
el movement, it varies from day to day.

Child's initial reaction tO most new foods, solids, liquids or
vitamins is to accept them without much fussing.

Child is usually willing to be held and cuddled by strangers.
Infant generally appears happy upon waking up.

If playing with one toy, the infant does not usually become dis-
tracted by others.

Infant's times for liquid feeding are unpredictable--vary more than
1 hour.

Baby usually sits or lies quietly throughout the feeding interval.

Child's reaction to bath, whether she likes or dislikes it, is
mild and not very excited.

Child has no regular waking time from naps--varies more than 1/2
hour.

Usually infant readily accepts new activity like swinging or using
a jumper.

When child is with one person, she/he will easily go to another
person.

During diapering and dressing, child's expressions are mild--
little smiling or fussing.

The infant shows discomfort with changes of place and situation
even after continued exposure.

Child will not crawl across room tO another toy if there are other
toys nearer tO him/her.

Child is generally happy when left alone in a room; she will occu-
py herself.

The infant Often continues playing no matter what does on around
him.

The infant usually whimpers or fusses slightly when hungry, but
does not cry loudly.

Child is a heavy sleeper; it takes a loud noise to wake him/her.
Child usually enjoys company.

Baby reacts tO an undesired food in a mild way.

446.

456.

466.

476.

486.

496.

506.

516.

526.

53C.

546.

556.

566.

586.

596.

606.

616.

626.

63G.

646.

656.

73

Child notices and reacts to small amounts Of urine in diaper.

When crying from getting a shot, infant can easily be calmed by
milk, pacifier, etc.

If bath is given in new place, infant readily accepts change.

Child has a negative reaction to all new foods, whether he/she
eventually accepts them or not.

Child protested when put into bath for the first time.
Child will fall asleep at about the same time each night.

During feeding, the child will continue to suck even if there is
much activity around him/her.

If left on the floor, infant will usually move to another area.
When napping, baby almost always sleeps through without waking.
Infant is active or playful on a fairly regular schedule.

Within a short period Of time, the infant generally accepts and
plays with a new toy.

Before going to sleep, child is Often fussy.
When food is Offered the infant does not attempt to reach for it.

Infant's general reaction to familiar peOple is intense--crying
Or laughing.

Child will not persist in activities that have previously upset
him.

Child can be left on couch or chair for periods of time without
moving very much.

Child cries for a long time after a frightening event is over.
Baby usually does not accept company (visitors).
Child liked his/her first tub bath.

During diapering and dressing, child is generally pleasant and
smiling.

During milk feedings child is not easily distracted and continues
to suck undisturbed.

Light or sound will at least temporarily stop the child's crying.

666.

676.

686.

696.

706.

716.

726.

736.

746.

756.

76C.

776.

786.

796.

806.

816.

fi

826.

74

Child does not seem to mind changes in amounts, kinds, or tastes
Of solid foods.

Baby Often consumes close to the same amount Of food during a
feeding.

Infant's general reaction to familiar people is mild--frown or
smile.

When infant is full, he/she shuts mouth, spits out food, or will
accept nothing else.

Child generally indicates that he/she has soiled.

When given a toy, the infant plays with it for many minutes.
Infant's play involves much movement and exploration.

After getting a shot, child cries for less than 1 minute.
Infant is easily distracted during breast or bottle-feeding.
Baby usually seems happy during bath.

Infant does not like to be bathed by different people.

Child will continue to engage in activities which have previously
caused an upset.

If baby's initial reaction to new procedure (nail cutting, hair
brushing, washing, etc.) is negative, acceptance usually takes
place after 2 or 3 times.

Baby drinks an unpredictable amount Of milk; (if bottle fed, var-
ies more than 2 02.; if breast fed, time sucking varies greatly).

Child is not very active during play. He diSplays few movements
and makes few sounds.

Infant has high tolerance for pain.

While playing with one toy, child can easily be distracted by

(f , another.

836.

846.

856.

86G.

Infant explores very little; needs help tO find play Objects.
Child is uninterested in eating at the same time each day.

He/she definitely notices and reacts to physical differences in
familiar people such as glasses, hats, etc.

Child will rarely allow strangers tO hold or cuddle him.

876.

886.

896.

906.

916.

926.

936.

946.

956.

966.

976.

986.

996.

1006.

1016.

1026.

1036.

1046.

1056.

1066.

1076.

75

If child wants a toy that is on the other side Of the room, he
will crawl until she reaches it; or' if unable to crawl, child
will continue to show an interest in it for quite a while.

During feeding time, child quickly pulls away from nipple when
milk is gone.

Child protested considerably to first bath.
When going tO sleep, infant is usually happy.

While playing, the infant is easily distracted by everyday occur-
rences like the ringing phone or doorbell.

When seeing a new animal, child's initial reaction is one of
interest and attraction.

Infant reacts tO slight temperature changes (in room or outside).

When given a food which he does not want, he reacts in a strong
manner--response is intense or powerful.

Infant exhibits regular, easily identifiable actions around meal
time.

Infant falls asleep at about the same time most nights--within
half an hour.

When being tickled, child laughs and seems to enjoy it.
When lying in crib, infant moves around a great deal.

Does not readily accept changes in types or characteristics of
foods.

Child continues to Object to grooming procedures (combing,
washing, nail cutting, etc.) even after experiencing them sev-
eral times.

Infant is usually fussy after feeding.

When in carriage or stroller, baby is usually quiet and still.

Infant's general reaction to familiar peOple is friendly with
laughing and smiling.

Changes in lighting will not stop the baby's crying.

Whether he likes or dislikes bathing, the infant's reaction is
usually intense or energetic.

When asleep, child actively moves around the crib.

When crying, child will persist until he/she is picked up.

1086.
1096.

1106.
1116.

1126.

1136.
1146.

1156.

1166.
1176.

1186.

1196.

1206.

1216.

1226.
1236.
1246.
1256.

1266.

1276.

1286.

76

When being washed or dressed, baby is generally pleasant, smil-
ing, etc.

Infant cannot occupy himself in crib or playpen for more than a
few minutes.

Until he gets the food he wants, baby will usually cry.
Infant initially accepts new foods.

When infant is full, he/she simply turns head away and lets food
drool out Of mouth.

While waiting to be fed, baby is very active and vocal.
If left on the floor, child rarely moves from spot.

If baby is hungry for milk, he will persistently refuse substi-
tutes such as pacifier, or water or juice.

Diapers must be heavily soiled before infant reacts.
Baby Often wakes during naps.

When introduced to new activity like swinging or using a jumper,
child is usually apprehensive.

Baby only reacts to large changes in room or outside temperatures.

At first, infant may reject a new toy but takes or plays with it
after several minutes.

When playing, baby will respond to hearing his/her name called.

After a number Of tries, the infant accepts changes Of place or
situation.

When child stops eating, playing a game or singing will get him to
resume eating a little longer.

Child is a light sleeper.
wake him/her.

It takes only the slightest noise to
Infant shows a mild reaction to light or sound with little or no
crying.

Child has a loud response to a wet or soiled diaper.

Child initially does not accept most new procedures; usually
cries, fusses or does not cooperate.

Baby usually fusses during bath.

1296.

1306.

1316 O
1326.

1336.

1346.

1356.

1366.

1376.

1386.

1396.
1406.
1416.

1426.

1436.
1446.
1456.
1466.

1476.

1486.

1496.

1506.

77

After a fall or bump infant cries for a long time.

Infant takes nap at approximately the same time each day--within
a half hour.

Infant is generally fussy during play.
Infant plays with one toy for only about 1-2 minutes.

Infant does not notice or react to changes in voice quality or
level.

When left alone for more than 5 minutes, child generally fusses
or cries.

After receiving a shot, it is difficult to stop baby's crying.

In playpen or on floor, infant is active; gets into things, pulls
at Objects, or puts nearby Objects in mouth.

When the lights are turned on in his/her room, child is easily
awakened.

When infant cries because of hunger, she will usually stop for at
least a minute if she is picked up, given a pacifier, etc.

Diapering is Often a battle.
Even after first 2 weeks Of bath, child continued to protest.
Does not follow a regular nightly sleeping pattern.

If there is any activity around him, child stops sucking during
feeding.

Infant can be fed at same time each day.

When she is hungry, almost nothing can make infant stop crying.
Child notices and reacts to changes in voice quality or level.
Infant notices and reacts tO slightly soiled diapers.

When there are interruptions in solid or milk feedings, the child
generally remains happy.

Diapers are usually very wet before baby shows any reaction.

Even after several trials, infant continues to reject most new
foods.

The baby will eat his meals at varying times during the week.

1516.

1526.

1536.

1546.

1556.

1566.

1576.

1586.

1596.

1606.

1616.

1626.

1636.

1646.

1656.

1666.

1676.

1686.

IF.

78

When receiving a shot, baby cries for several minutes or more.
Baby drinks a predictable amount Of milk; (if bottle-fed, varies
less than 2 02.; if breast fed, time sucking does not greatly
vary).

Infant usually lies still while being diapered or dressed.

Child cheerfully tries new foods whether he/she ultimately likes
them.

Initial reaction to strangers is relatively mild such as a frown
or smile.

When diaper is wet or soiled, it is difficult to distract baby
from fussing.

Child generally takes milk around the same time Of day; does not
vary more than 1 hour.

Child still exhibits strong reactions even after repeated con-
tacts with bright light or loud sound.

Child will usually stop crying on his/her own without picking
him/her up.

Infant usually plays quietly and calmly.

Infant does not become easily accustomed to changes in caretakers
--babysitters, grandparents, etc.

When sick, baby is usually himself with some whimpering but not
much crying.

Whether liking or disliking a food, baby's response to it is
dramatic.

If an Object is out Of reach, infant continues to reach for it for
several minutes.

If child is fussing about a soiled or wet diaper, he/she can eas-
ily be distracted for a few minutes.

Infant initially accepts any new procedure.

Sudden appearances of strangers will cause crying and/or a turn-
ing away.

Even when given a substitute infants will cry when something is
taken from her reach.

When being washed or dressed, infant generally cries or fusses.

2P.

3P.

4P.
5P.

6P.

7P.

8P.

9P.
10P.
11P.
12P.

13P.

14P.
15F.
16? O

17P.

18P.

19P.

20F.

21F.

22P.

23p 0

79

Although child initially protested bath, she accepted it after 2
or 3 times.

For at least 3-5 minutes, child will lie and watch a hanging
mobile.

Infant is usually cheerful during play; laughing, smiling, etc.
When hurt, he/she usually cries loudly for long periods Of time.

Infant generally cries when solid or milk feedings are inter-
rupted.

In a crib or play pen, infant can amuse him/herself for quite a
while.

If child fusses during washing and grooming activities, he/she is
easily distracted by games, toys, etc.

Baby fusses or cries when he/she is tickled.

Baby Often cries loudly when hungry.

In her stroller, infant is usually quite active or noisy.
If child falls, he/she does not cry for a long time.

Child makes himself at home most anywhere; appears comfortable
in new situations.

After infant has been fed, he/she usually smiles and laughs.
It is difficult to predict infant's activity or play time.
Child resists going tO different persons.

Easily notices and reacts to differences in taste and consist-
ency of food.

Within a few days, child adjusts to changes in familiar routines.

When brought to the doctor for a well baby check-up, child is
usually fussy.

Infant fusses when held down for diapering but he/she can usual-
ly be diverted with a pacifier or toy.

Baby adjusts easily to different care-takers.
If bath is given in new place, infant generally fusses and cries.

Child usually has a bowel movement at the same time Of day, with
less than one hour variance.

24F C
25P.

26P.

27F.

28P.

29P.

30P.

31p.
32P.
33P.

34P.

35P.

36P.

37F.

38F O

39P.

40P.

41P.

42P.

43P.

44P O

80

When left lying in crib, infant usually lies quite still.
If toy is taken from baby, he/she will fuss until it is returned.

Infant's reaction to animals is intense with much laughing or
crying.

While playing child does not notice or react to his/her name
being called.

Baby readily accepts bathing by a new or different person on
the first or second time.

When diaper is wet or soiled, child makes no fuss or whimpers
slightly.

Infant cannot be left for very long on couch or bed because he
might wiggle Off.

Sucks actively on breats, bottle, or pacifier.
The baby will resist being fed at varying times each day.
He/she frequently squirms or kicks during feedings.

Child wakes up from napping at approximately the same time every
day (within half an hour).

Baby persists in attempts to crawl or walk until a few steps are
taken.

Baby shows little reaction to animals (dogs, cats, etc.).

After 1-2 tries baby adjusts easily to changes in feeding sched-
ule.

The infant initially tolerates or enjoys new places and sit-
uations.

Infant will not fall asleep at the same time each night.

In response to pain, infant whimpers or cries, for only a short
time.

Child stops resisting milk substitutes such as water, juice, or
pacifier within a few minutes and accepts what is Offered.

Even after several minutes, infant continues to reject a new toy.
Child seemed tO enjoy his first car ride.

When lying in crib, child tends to focus on one Object for about
5 minutes at a time.

45P.

46P.

47P.

48P.

49P.

SOP.

51P.
52P.

53P.

54P.

55P.

56P.

57P.

58P.

59P.

60P.

61F.

62P.

63P.

64F.

81

Infant does not adjust easily to efforts at changing feeding
schedule.

Baby shows little reaction to bright lights or loud noises.

If fussy during grooming procedures (nail cutting, hair brush-
ing, etc.), baby is not easily distracted.

Child initially reacts to strangers with much laughing or crying.

Infant will readily accept bathing by a different person or in a
different place.

Child seldom seems to notice or react to differences in the taste,
consistency, or temperature Of foods.

Changes in sound (voices, TV, radio) will not stop baby's crying.
Child usually rejects new foods.

When trying to turn over or crawl, child tries for a minute or
two, then gives up.

Infant takes 3 or more days to adjust to changes in daily sched-
ule.

Baby's first reaction to a new animal is often accompanied by
hesitation or fear.

Infant's time Of waking is not consistent from day to day (times
vary more than half an hour).

Child will usually lie and watch a hanging mobile for just a
short period Ot time (30 seconds or less).

After repeated exposure to bright light on a short period Of time
or loud sound, child's reaction lessens.

At the doctor's for a well baby check-up infant is usually
friendly and smiling.

Baby smiles, gurgles, or plays with new people.

Baby shows no easily identifiable sign Of hunger when meal time
arrives.

Child seemed to dislike his/her first car ride.

Infant does not readily tolerate or enjoy new places and situa-
tions.

He/she mildly smiles or coos when served a desired food.

82

65P. Child has low tolerance for pain.

66F. Baby lies fairly still while he/she sleeps.

67P. Child seldom or never indicates that diaper is wet.
68F. Child continues to suck even after milk is gone.
69F. Baby strongly reacts to light or sound.

70?. Infant has no regular time pattern for napping each day (varies
more than 1/2 hour).

71P. While playing, infant generally does more quiet observing than
active exploring.

72?. When engaged in play, baby is usually actively moving and making
sounds.

73P. Child usually indicates that diaper is wet.

74?. When given a food, he/she does not like, infant protests briefly
but soon takes it anyway.

75P. Infant usually stops trying for a toy out of reach in less than 1
minute.

76F. Child seldom or never indicates that he/she has soiled (b.m.).

83

Table 14

Scoring Key for Michigan Infant Temperament Scale

Scale

Activity
Adaptability
Intensity
Threshold
Persistence
Mood

Approach
Distractibility

Rhythmicity

Items

6, 8*, 14, 21, 30*, 51, 56*, 59*, 72, 80*, 98, 102*,
106, 113, 114*, 136, 153*, 179, 192*, 198, 199,
201, 234, 239*, 240

9, 34, 36*, 46, 66, 76*, 78, 99*, 100*, 120, 122,
139*, 140*, 149*, 158*, 161*, 170, 184*, 186, 189,
190*, 196, 205, 210*, 213*, 217, 222*, 226*

11, 18, 31*, 35*, 40*, 43*, 57, 68*, 69, 73, 94*,
105, 112*, 125*, 126, 151, 155*, 160*, 173, 178,
194, 197*, 204*, 208*, 216, 232, 237*

12, 20, 23*, 41, 44*, 70*, 81, 85*, 93*, 116*, 119,
124*, 133*, 137*, 145*, 146*, 148, 181, 185*, 214,
218, 233*, 235, 241*, 245

1*, 3*, 37*, 39, 58*, 60, 71, 77, 87, 88*, 106, 110,
115, 129, 132*, 159*, 162, 164, 169*, 171, 175,
180*, 193*, 203, 209*, 221*, 225*, 236, 242*, 243*

2*, 5*, 7*, 22*, 27, 38, 47*, 55*, 63, 75, 90, 97,
101*, 103, 108, 128*, 131*, 134*, 147, 154, 172,
174*, 177*, 182, 187*, 227

4*, 10, 25, 26, 33, 42, 48*, 54, 61*, 62, 83*, 86*,
89*, 92, 111, 118*, 127*, 166, 167*, 206, 211, 220*,
223*, 228, 230*, 231*

13, 15, 19, 28*, 45, 50*, 64*, 65, 74, 82, 91, 104*,
121, 123, 135*, 138, 142, 144*, 156*, 165, 168*,
176, 188, 195*, 212*, 215*, 219

16, 24*, 29*, 32*, 49, 52, 53, 67, 79*, 84*, 95, 96,
117*, 130, 141*, 143, 150*, 152, 157, 183*, 191,
200*, 202, 207*, 224*, 229*, 238*

Scoring: Items marked 'T' are to be scored 1 and items endorsed 'F'

scored 0.

Starred items (*) are to be scored in reverse,

i.e., F = 0 and T 8 0.

M.I.T.S. Supplementary Information Form

First Name of Infant

 

What was the birth weight of your child? lbs. oz.

What was the expected delivery date for him/her?
(What your obstetrician told you) month day year

 

List all the children in your family (including the infant listed above)
in order of age.

 

 

 

 

 

 

1. yrs mos. M F (circle) 4. gyrs mos. M F (circle)
Sex Sex

2. gyrs mos. M F (circle) 5. yrs mos. M F (circle)
Sex Sex

3. yrs mos. M F (circle) 6. gyrs mos. M F (circle)
Sex Sex

What kind of job does the father in the family have? (Please list job
title and give a brief description)

 

job title

Description:

 

Highest school grade completed by father:

1 2 3 4 5 6 7 8 9 10 11 12 1 2 3 4 Some graduate Graduate
1 College training degree
What kind of job do you do? (if housewife, please list)
job title

 

Description:

 

Highest school grade completed by mother:

1 2 3 4 5 6 7 8 9 10 11 12 l 2 3 4 Some graduate Graduate

College training degree
What is the average amount of time per week day each parent spends with
the infant (the child you filled out the questionnaire on)?

Father: % 1 1% 2 2% 3 3% 4 4% 5 5% 6 hours or greater

Mother: % 1 1% 2 2% 3 3% 4 4% 5 5% 6 hours or greater
What is the average amount of time per weekend day each parent spends
with the infant?

Father: % l 1% 2 2% 3 3% 4 4% 5 5% 6 hours or greater
Mother: % l 1% 2 2% 3 3% 4 4% 5 5% 6 hours or greater

84

85

The following questions ask for somewhat more detailed information, but
we hope you will answer them. All information will be kept completely
confidential; no one other than the researchers will have access to any
of it. In addition, all identifying names are deleted before we work
on it.

1.

2.

Does your child attend a day-care center? yes no (circle)

Is your child cared for by someone other than an immediate family
member (father, mother, or sibling) for three (3) or more hours per
week?

yes no

If your answer was yes to questions 1 or 2, may we contact the care-
taker or teacher to ask them to fill out a questionnaire like the
one you just completed? No results or information will be given to
this person other than that you have been participating in this
study and have consented to our contacting them.

yes no

If yes to #3, list the name of the caretaker or day-care center that
takes care of your child.

 

Would you be willing to be recontacted for a follow-up to this
questionnaire? yes no

Are you interested in learning more about this study? yes no
If yes, we will send to you within the next 3 to 4 months a sum-
mary of the purpose and the overall (group) results of the study.

Comments:

ALTHOUGH IT IS OPTIONAL, IF YOU HAVE ANSWERED YES TO ANY QUESTION
ON THIS PAGE, PLEASE LIST YOUR NAME AND ADDRESS BELOW. (As noted
above, we need this so that we can recontact you at a later date.)

 

Name

 

Number and Street

 

City State Zip

 

Telephone Number

APPENDIX C

APPENDIX C

GENERAL INFORMATION:
Dear Parents:

Thank you for participating in our study of infant behavior. Enclosed
you will find a questionnaire packet designed to gather information from
you, a knowledgeable informant on your infant's behavior.

The questionnaire should take you no more than 1 hour to complete. In
fact, most individuals require 45 minutes or less. It is extremely
important that you do the questionnaire in one sitting. In other words,
wait until you have 45 minutes of time where you are unlikely to have
many distractions or responsibilities that would compete with complet-
ing the forms. Many parents find that naptime or after the children are
in bed to be convenient times for this. Please do the questionnaire
alone, without consulting others on the answers to any of the items.

You should have two separate forms and an addressed, stamped envelope

in the packet sent to you. These forms are: Departmental Research Con~
sent Form, and M.I.T.S. questionnaire. Please complete these forms in
the following order and in exact accordance with the instructions.

GO TO THE DEPARTMENTAL RESEARCH CONSENT FORM

1. Completely read the consent form. This form indicates that you are
voluntarily participating in the study. Please read and sign your
name and record the date at the bottom of this page. All identify-
ing information will be removed from the forms and the results will
be held completely confidential. By doing this work now, you are
under no obligation to participate in the follow-up aspects of the
study.

NOW TURN TO THE QUESTIONNAIRE (M.I.T.S.)

2. Now, carefully read the instructions on the front of the M.I.T.S.
questionnaire. When you have finished reading the instructions,
begin the questionnaire. Read each item and put your answer (T or
F) to the left of the item number. Remember, do the questionnaire
alone. Do not seek any help. When you finish the questionnaire,
put it in the return envelope.

When you are done, put the forms in the return packet, and mail it back
to us. The envelope is pre-addressed and stamped so you need only place
it in a local mailbox. If you have requested information about the
study and/or expressed interest in participating in a follow-up ques-
tionnaire, we will contact you within the next 3 months. However, if
you have any questions, contact us at any time at the address above, or
by phone (Mr. Bonem) at 355—9564 (8 to 5) or evenings at 332-2486.
Again, thanks for your cooperation.

86

APPENDIX C

GENERAL INFORMATION letter
Page 2

Sincerely,

Howard Bonem

HB/Rz:sh

87

Robert Zucker, Ph.D.

INSTRUCTIONS

You should have the M.I.T.S. questionnaire and an addressed, stamped
envelope.

1.

3.

Put today's date on the top of the M.I.T.S. questionnaire. This is
very important as we need to know the length of time since you last
took the instrument.

Carefully read the instructions on the front of the M.I.T.S. ques-
tionnaire. When you have finished reading the instructions, begin
the questionnaire. Read each item and put your answer (T or F) to
the left of the item number. When you finish the questionnaire, put
it in the return envelope.

Please mail the completed form back to us. If you have any ques-
tions, give us a call.

Thank you for your help.

88

APPENDIX D

APPENDIX D

MICHIGAN STATE UNIVERSITY
Department of Psychology
Snyder Hall

East Lansing, Michigan 48824

Dear Father:

Your wife has already taken part in a research project directed towards
studying the behavior of infants 3 to 12 months of age. Our interest

is in the area of styles or patterns of infant behavior. We are con-
tacting you now to request your participation in the study also by fil-
ling out a questionnaire. This questionnaire has true-false questions
about things an infant does and takes about 30 to 40 minutes to complete.
It is important for us to see how fathers judge their infant's behavior
as well as mothers. In order to do this, we would like you to complete
the enclosed questionnaire according to the attached instructions.

Please note that your wife's participation in this project in no way
obligates you to participate in this study. However, we hope you will
choose to do so. Your answers will remain completely confidential:

all identifying information will be removed from the forms before they
are analyzed. Your answers will be quite helpful to us and will help
give us a better understanding of styles of infant behavior.

If you have any questions, feel free to contact us at the address above,
or by phone (Mr. Bonem) at 355-9564 (8 to 5) or evenings at 332-2486.
Thank you for your cooperation.

Sincerely,

Howard Bonem

Robert A. Zucker, Ph.D.

HB/RAZ:sh

89

GENERAL INFORMATION:
Dear Parents:

Thank you for participating in our study of infant behavior. Enclosed
you will find a questionnaire packet designed to gather information
from you, a knowledgeable informant on your infant's behavior.

The questionnaire should take you no more than 1 hour to complete. In
fact, most individuals require 45 minutes or less. It is extremely
important that you do the questionnaire in one sitting. In other
words, wait until you have 45 minutes of time where you are unlikely
to have many distractions or responsibilities that would compete with
completing the forms. Many parents find that naptime or after the
children are in bed to be convenient times for this. Please do the
questionnaire alone, without consulting others on the answers to any
of the items.

You should have two separate forms and an addressed, stamped envelope
in the packet sent to you. These forms are: Departmental Research
Consent From, and M.I.T.S. questionnaire. Please complete these forms
in the following order and in exact accordance with the instructions.

GO TO THE DEPARTMENTAL RESEARCH CONSENT FORM

1. Completely read the consent form. This form indicates that you are
voluntarily participating in the study. Please read and sign your
name and record the date at the bottom of this page. All identi-
fying information will be removed from the forms and the results
will be held completely confidential. By doing this work now, you
are under no obligation to participate in the follow-up aspects of
the study.

NOW TURN TO THE QUESTIONNAIRE (M.I.T.S.)

2. Now, carefully read the instructions on the front of the M.I.T.S.
questionnaire. When you have finished reading the instructions,
begin the questionnaire. Read each item and put your answer (T or
F) to the left of the item number. Remember, do the questionnaire
alone. Do not seek any help. When you finish the questionnaire,
put it in the return envelope.

When you are done, put the forms in the return packet, and mail it
back to us. The envelope is pre-addressed and stamped so you need
only place it in a local mailbox. If you have requested information
about the study and/or expressed interest in participating in a
followbup questionnaire, we will contact you within the next 3 months.
However, if you have any questions, contact us at any time at
332—2486. Again, thanks for your cooperation.

Sincerely,

Howard Bonem Robert Zucker, Ph.D.
90

APPENDIX E

APPENDIX E

'MICHIGAN STATE UNIVERSITY
Department of Psychology
Snyder Hall

East Lansing, Michigan 48824

Dear

We are presently engaged in a study of infant behavior. Our interest
is in the area of patterns or styles of infant behavior. In order to
measure this, we have devised the enclosed questionnaire (a 244 item,
true-false scale).

has been involved with our study and has already filled out
a questionnaire on her child. She has told us that you often care for
this child and therefore are knowledgeable about this child's behavior.

She has also given us permission to contact you to see if you would
also fill out this questionnaire on This will allow us to
compare the ratings of two knowledgeable informants, one a mother and
the other a more neutral caregiver. Although you are under no obli-
gation to participate, we hope you will choose to do so. Your answers
will remain completely confidential with all names and identifying
information being removed from the forms before the answers are anal-
yzed.

We hope you will be able to cooperate. Please carefully read the
enclosed instruction sheets before completing the forms. If you have
any questions, feel free to contact us. You may call Mr. Bonem at
355-9564 (8:00 - 5:00) or in the evening at 332-2486.

Thank you for your help.

Sincerely,

Howard Bonem

Robert A. Zucker, Ph.D.

P.S. A photocopy of Mrs. release form is enclosed for your
inspection.

91

MICHIGAN STATE UNIVERSITY
Department of Psychology
Snyder Hall

East Lansing, Michigan 48824

GENERAL INFORMATION:
Dear Caregiver:

Thank you for participating in our study of infant behavior. Enclosed
you will find a questionnaire packet designed to gather information
from.you, a knowledgeable informant on behavior.

The questionnaire should take you no more than 1 hour to complete. In
fact, most individuals require 45 minutes or less. It is extremely
important that you do the questionnaire in one sitting. In other words,
wait until you have 45 minutes of time where you are unlikely to have
many distractions or responsibilities that would compete with complet-
ing the forms. Many parents find that naptime or after the children

are in bed to be convenient times for this. Please do the questionnaire
alone, without consulting others on the answers to any of the items.

You should have two separate forms and an addressed, stamped envelope

in the packet sent to you. These forms are: Departmental Research Con-
sent Form, and M.I.T.S. questionnaire. Please complete these forms in
the following order and in exact accordance with the instructions.

GO TO THE DEPARTMENTAL RESEARCH CONSENT FORM

1. Completely read the consent form. This form indicates that you are
voluntarily participating in the study. Please read and sign your
name and record the date at the bottom of this page. All identify-
ing information will be removed from the forms and the results will
be held completely confidential. By doing this work now, you are
under no obligation to participate in the follow-up aspects of the
study.

NOW TURN TO THE QUESTIONNAIRE (M.I.T.S.)

2. Now, carefully read the instructions on the front of the M.I.T.S.
questionnaire. When you have finished reading the instructions,
begin the questionnaire. Read each item and put your answer (T or
F) to the left of the item number. Remember, do the questionnaire

alone. Do not seek any help. When you finish the questionnaire,
put it in the return envelope.

When you are done, put the forms in the return packet, and mail it
back to us. The envelope is pre-addressed and stamped so you need
only place it in a local mailbox. If you have any questions, con-
tact us at any time at the address above, or by phone (Mr. Bonem)

at 355-9564 (8 to 5) or evenings at 332-2486. Again, thanks for your
cooperation.

92

GENERAL INFORMATION letter
Page 2

Sincerely,

Howard Bonem

HB/RZ:sh

93

Robert Zucker, Ph.D.

Michigan State University
Department of Psychology

DEPARTMENTAL RESEARCH CONSENT FORM

I have freely consented to take part in a scientific study being
conducted by: Howard Bonem under the supervision of:
Dr. Robert A. Zucker , Academic Title: Professor .

 

The study has been explained to me and I understand the explan-
ation that has been given and what my participation will involve.

I understand that I am free to discontinue my participation in
the study at any time withOut penalty.

I understand that the results of the study will be treated in
strict confidence and that I will remain anonymous. Within these
restrictions, results of the study will be made available to me
at my request. ,

I understand that my participation in the study does not guaran-
tee any beneficial results to me.

I understand that, at my request, I can receive additional explan-
ation of the study after my participation is completed.

Signed

 

Date

94

APPENDIX F

95
APPENDIX F
Table 15

Items and their Loadings on each Persistence Scale Factor,

and Coefficient Alpha and Item-total Correlations for each Factor

 

 

Coefficient Range of Item-total

Factor Loadings Alpha Correlations

Factor 1 .72 .25 - .54
Items: 3 .61
37 .21
87 .42
164 .49
203 .45
221 .65
243 .55

Factor 2 .66 .19 - .52
Items: 132 .29
169 .21
171 .69
175 .52
225 .69

Factor 3 .64 .24 - .64
Items: 107 .71
59 .89
162 .21

Factor 4 .28 .02 - .35
Items: 88 .69
221 .20
236 .83
242 —.31

Factor 5 .56 .03 - .35
Items: 58 .73
77 .82
110 .25

Factor 6 .65 .48

Items: 115 .70

209 .76