JHESLS

['0
«I
'I

This is to certify that the

thesis entitled

presented by

has been accepted towards fulfillment

of the requirements for

 

  

Major protessor

 

Date fiZW /é /fl,;t3

.‘ /I

//

L/

0-169

 

 

A TECHNIQUE FOR THE DESCRIPTION OF THE SELF

BY

Lawrence Walker

A THESIS

Submitted to the School of Graduate Studies of Michigan
State College of Agriculture and Applied Science
in partial fulfillment of the requirements

for the degree of

MASTER OF ARTS

Department of Psychology

5 Year 1953

thsas

ACKNOWLEDGEMRNTS

The author is indebted to Dr. D. M. Johnson
under whose guidance and sympathetic interest
this study was undertaken. '

The author is also grateful to Dr. H. A. Grace
and Dr. A. G. Dietze who read and criticized the

manuscript.

\t"

‘3 a" ‘ 5“,," “‘ 43..."! 3
{RAJLMJL

TABLE OF CONTENTS

Page
INTRODUCTION. . . . . . . . . . . . . . . . . . . . . . 1
SOME MODERN TECHNIQUES. . . . . . . . . . . . . . . . . 5
PURPOSE . . . . . . . . . . . . . . . . . . . . . . . . 9
TECHNIQUE AND METHOD. . . . . . . . . . . . . . . . . . 1E
RESULTS . . . . . . . . . . . . . . . . . . . . . . . . 18
SUMMARY AND CONCLUSIONS . . . . . . . . . . . . . . . . 35
REFERENCES. . . . . . . . . . . . . . . . . . .|. . . . 39

APPENDIX. 0 O O O O O O O O O O O O O O 0 O O O 0 O O 0 LL]-

LIST OF TABLES
Table Page
I Split-half reliability coefficients for
each category corrected by the Spearman-
Brown prOphecy formula. . . . . . . . . . . . 19

II Intellectual, Physical, Home, and Group
categories and their intercorrelations. . . . 19

III Mean ratings for each category. . . . . . . . . 22

IV 't' ratios between the means of the Intell-
ectual, Physical, Home, and Group
categories. 0 o c o o o o o o o o o o o o o o 22

.<

Mean ratings for each category by sex . . . . . 2h

VI Mean ratings for each category by class
levels. 0 o o o o o o o o o o o o o o o o o o 27

VII Mean ratings for each category by age levels. . 27

LIST OF FIGURES
Figure Page

1 Mean ratings for each category by class
levels. 0 o o o o o o o o o o o o o o o o o o 28

2 Mean ratings for each category by age levels. . 29

INTRODUCTION

The term 'self' is used in many different contexts.
Historically the term evolved from a Spiritual heritage
embodied in the concept of the 'soul' which was supposed to
account for the actuating cause and purposiveness of mental
life. Under the impact of Wundt and the positivists the
term 'soul' fell into disrepute among psychologists. James

5?11) and Calkinscfh) were largely responsible for calling
attention to the 'self' and making it a pOpular concept.

Since late in the nineteenth century, the term 'self'
has been bandied about among psychologists so that today
it is vested with an overflow of meaning. Epistemologists
deal with a 'self' or 'subject' which knows objects. Modern
psychotherapists view the 'self' or 'ego' as an entity to
be diagnosed and treated. Many psychologists have assumed
some sort of active agent serving a cognitive function such
as intending, knowing, etc. Others have treated the 'self'
as an abstraction or theoretical construct outside the realm
of eXperience and direct description. For some the 'self'
has served the dual function of knower and object of knowledge.

The structuralists and behaviorists generally avoided
the concept, although Titchener and his associates trained
their introspective sights on the 'self' and found nothing

but kinesthetic feelings and sensations. Brentano and the

'act' psychologists maintained that conscious acts implied
the 'self' as actor and as object although the 'self—object'
was of secondary importance. Calkins (h) and the person-
alists believed that the 'self—object' was the same as the
active self. James (11) discussed the 'empirical self'
as an object in the stream of consciousness. He said that
"in its widest possible sense, however, a man's Self is the
sum total of all he CAN call his." (p. 291). Thus the 'self'
can be divided into material, social, and Spiritual selves
and the pure ego. James' selves were the products of psycho-
logical performance at times, but were often alluded to as
an object in consciousness rather than as an active agent.

Failure to distinguish clearly between the 'self' as
doer and the 'self' as object has led to the confusion that
now exists. Hilgard (9) has described this approach in
these words:

". . . you presently find yourself as between two

mirrors of a barber shOp, with each image viewing

each other one, so that the self takes a look at

itself, taking a look at itself. It soon gets

confused as to the self that is doing the looking

and the self which is being looked at." (p. 377)-
In summarizing the literature on the ego, Allport (2)
enumerated eight capacities of the ego, viz., (l) as knower,
(2) as object of knowledge, (3) as primordial selfishness,
(h) as dominator, (S) as a passive organizer and ration-
alizer, (6) as a fighter for ends, (7) as one segregated

behavioral system among others, (8) as a subjective pattern-

ing of cultural values (p. #59). He viewed the 'self' as

a knower, organizer, observer, status seeker, and as a
socialized being. In his theorizing Allport posited ego-

involvement as a ". . . condition of total participation

 

of the self. . ." (p. h59). He then cited experimental
evidence to buttress his position and concluded:

"In the experiments I have cited, and in many
others of an analogous nature, it turns out that
one group of subjects (those who are personally
aroused and committed to a task) behave in ways
quite unlike other subjects (who are not so
committed). . . . In short, we are here con-
fronted with some parameter that makes a vast
difference in our eXperimental results." (p. #72).

Although Allport presented a convincing case for the

importance of ego-involvement as a determiner of behavior,

 

he admitted that the concept of 'self' or 'ego' has not
been adequately defined.

Koffka\%l2) and Lewin (I3) regarded the 'self' as a
particular field part in constant interaction with the rest
of the field. The self was conceptualized as a sub-system
of the ego which is in various degrees of interaction with
the other sub-systems wherein certain tensions arise if the
ego becomes engaged. Both agreed that the facts of action
are the empirical foundation of the ego. Snygg and Combs
(21) who have been greatly influenced by the Gestaltists
Spoke of a 'self—concept' that included all those aspects
of the field which the individual perceives to be definite
and fairly stable characteristics of himself. Rogers (18)
who eSpoused a similar view said that "by this 'concept of

self' we mean the individual's perceptions of his own

characteristics and his relations to others, and the values
he attaches to these perceptions" (p. 68).

Although this historical sketch is by no means compre-
hensive, it Should be clear that there is a pceponderance
of seemingly disconnected theories which purport to deal
with the same phenomena. A renewed interest to the problem
of the 'self' or 'ego' has taken a firm foothold in eXperi-
mental psychology. With their modern tools and techniques
the scientifically oriented psychologists have begun the
weighty task of proving the validity or worthlessness of
the morass of speculation surrounding the 'self', 'ego',
‘self—concept', etc. Although many experiments in the area
of personality are related to the 'self', only those

focusing particularly on the 'self' will be examined.

SOME MODERN TECHNIQUES

The techniques employed in the studies of the 'self'
and 'self-concept' are varied and ingenious. Personal
documents, behavior in controlled and uncontrolled situations,
projective techniques, personality schedules, and question—
naires have provided sources for studies concerning the
'self-concept'. In all cases, however, the nature of the
'self' is interpreted from palpable data which provide the
basis for inference.

The recording of therapeutic sessions at the Counseling
Center of the University of Chicago has provided a rich
source of data from which has develOped a great deal of
research. Sheerer (l9) carefully defined "self-acceptance"
and "acceptance of other persons". With these definitiOns
four judges rated on a five-point scale 51 recorded state-
ments showing some self evaluation and 50 statements reveal-
ing evaluative attitudes toward others. The scale accounted
for degrees of acceptance of self and others. Among other
things the study suggested that acceptance of self is
definitely related to acceptance and respect for others.
Stock (23) employed a similar procedure, but the statements
were judged for intensity and direction of feeling.

Philips (16) utilized a ten-item questionnaire with five

items relating to self and five items relating to others.

In this method the individual himself rated the statements
as to how applicable they were to him. This much simpler
instrument yielded about the same general results as the
aforementioned method of protocol analysis.

The problems of insight and stability of the self-
concept have recently found their way into experimental
psychology. Rogers (18) expounds and uses a method called
the Q technique which compares a client's internal reference
of himself with the external reference of a psychologist
toward the patient. The assumption underlying Roger's
method is that insight, operationally defined as the degree
of agreement between the internal and external frames of
reference, should increase as therapy progresses. The
internal reference is obtained by the technique of self-
rating. One hundred self-descriptive statements relating
to behavior or internal states are sorted into nine piles
and the patient is asked to arrange the statements according
to how closely he thinks they describe him. He is also
asked to re-sort the cards to represent the person he would
like to be, his ideal self. The patient is given the
Thematic Apperception Test which a psychologists evaluates
blindly. The psychologist then arranges the one hundred
statements according to the personality picture as revealed
by the Thematic Apperception Test. Ideally, at the end of
therapy the evaluation of the diagnostician should agree
with the individual's self-evaluation, and both should agree
with the ideal self.

Holt (10) and Grossman (6) reported ways by which
insight may be determined. The former eXperimenter used a
method whereby individuals rated themselves as regards
35 needs, and then experts who had interviewed them gave
their ratings for each individual. The differences between
the self-ratings and the experts' ratings yielded a measure
of insight. Grossman (6) constructed two tests of insight.
The first measure used a discrepancy score between self-
evaluations on a personality schedule and self-evaluations
on specific traits derived from the personality sChedule.
The individual was instructed to rate himself as compared
to his peer group. The other technique involved the con-
struction of multiple-choice items to cover the attitudes
revealed by the subjects on Thematic Apperception Cards.
According to Grossman the measure of insight was:

"The degree of discrepancy between the manner in

which the subject answered the items and the

manner in which the two psychologists felt he

should have answered it ... indicated the amount

‘ET—IHEight the subject possessed (p. 111).

Brownfain (3) investigated the stability of the self-
concept. The subjects rated themselves on 25 personality
variables. Under one set of instructions the individual
was told to give himself any benefit of a doubt, and then
the second set of instructions required the subjects to re-
rate themselves but this time they were told not to give
themselves the benefit of a doubt. The absolute difference
between the two self pictures was a measure of the stability

of the self-concept.

The recent trend has been in the direction of ratings
and evaluations by self and others in reSpect to feelings,
attitudes, and remembered or anticipated behavior as they
relate to the individual. The techniques and instruments
appear to be as varied as there are psychologists working
on the problem. The potential methods and experimental
designs in this area are limitless, but researchers employ-
ing different techniques have already approached essential
agreement (6, 10, l6, 19, 23).

PURPOSE

In the midst of the confusing philOSOphizing and
Speculation concerning the 'self', there is one general facet
of the 'self' which has been universally agreed upon either
directly or by implication. But for some inexplicable
reason experimenters and researchers have neglected this
aspect of the 'self'. The thread of continuity runs through
the writings of innumerable authors since the time of William
James (1, 2, 11, 12, 13, 15, 17, 20, 22). This common
position although stated in language unique to each author
may be summarized thusly:

A. The subjective, phenomengl, or perceived self is
not limited by_the boundaries of the body. The evidence
points to the notion that one's own body with its appurten-
ances, family, friends, country, culture, race, social and
class groups, etc. can be in fact an integral part of the
self. Even the values which are inculcated by virtue of
being a member of a group may be more important than the
body itself. The extensions or projections which refer to
the self might be termed self-referents. These self-referents

are organized into areas forming the self-structure. The

 

boundaries of these self-areas though variable might be
charted and delineated. The catatonic schiZOphrenic who

has contracted his boundaries to nothing save the pulsing

10

of his heart is representative of one extreme, while the
mystic who has attained the state of "cosmic consciousness"
portrays the other extreme. Both of course are pathological,
but they are poignantly illustrative. Literature too is
fraught with perceptive observations regarding the solu-
bility and extensions of the self.

It seems then that anything an individual holds dear
whether it be objects, persons, values or abstractions and
which can be considered as me, my, or mine are to a certain
degree functionally equivalent and are rooted in the sub—
jectively felt self. Any threat or disapproval directed
toward these self-referents is perceived as danger or
insult, and enhancement or approval of these self-referents
is perceived as gratifying. The prOposition presents itself
that instigations to self-defensive or self-enhancing be-
havior are dependent upon the structure of these self-
referents as it interacts with forces which are either
threatening, approving or neutral.

B. The subjective,_phenomenal, or’perceived boundaries

 

separating the self-referents into areas tend toward con-
sistency but are nonetheless modifiable. From a genetic
point of view, as a child matures, the self-referents and
their boundaries encompass more and more psychological
territory until a point is reached in adulthood when the
bounds become relatively fixed and stable. Adolescence

marks the climax of this reorganizing process. When the

ll

organization and deveIOpment of these self-referents are
determined, predictions can be made and hypotheses can be
tested.

The impact of cultural and social forces upon the
individual and the reinforcing nature of his membership
therein determine to a great extent a basic and common
self-structure. The process of social interaction and its
consequent identifications, though, determines differential
self-structures and self-referents which are related to
culturally defined roles. Thus there are probably both
similarities and differences in the self-structures of males
and females. In fact there are probably as many different
areas of the self as there are definite groups with which
an individual identifies although these are imbedded in a
more basic cultural matrix.

These two general prOpositions suggest that some
technique should be devised which would Open this area to
experimental verification. Specifically, the object of the
present study was to investigate the feasibility of a
technique which would chart the unexplored regions of the
extended self and its referents. The technique decided upon
was a self-rating schedule. Although this is a simple
instrUment, it was felt that the merits of a straightforward
device should be investigated first before attempting to use
the complex projective methods. Accordingly, then, the

primary aim was to demonstrate whether an instrument which

12

meets the necessary statistical requirements can be deveIOped.
Since the reliability of such an instrument is probably

the sine qua non of its validity, the former problem logically

 

must be settled first. When this is accomplished, the
inductive-deductive process of science with its strictures
of validity and empirical verification can begin.

A secondary intent of this study was to pose some
questions that follow from theoretical considerations in
an effort to partially answer them. Of course the reli-
ability of this instrument must be assumed if meaningful
answers are to be eXpected, but the work of Hartocollis (8)
and Gladin (5) indicated that this would be the case. It
was hoped that this method would shed light on the following
questions.

1. Are there distinguishable areas of the self—

structure differing in psychological signifi-
cance to the individual?

Since this is a nomothetic approach, and since a
select pOpulation of college students £55 tested, the
answer to this and the following questions applied only to
groups and more particularly college groups. The self-
structures of college students have much in common. Tradi-
tionally the college pOpulation is composed of a select group
primarily representing a certain segment of American society
who are thrown together in a common social milieu. Further,
selective factors such as native intelligence, financial

security, and class eXpectations operate in the direction of

13

attracting and molding certain characteristic types of
self-structures. Therefore, it seems likely that intel-
lectual, college, and home referents might tap different
crucial areas of the self structure. A fourth area util-
izing physical self-referents was deemed important since

the American college culture places a high value on physical
appearance.

2. Do the sexes exhibit differences in their self-
structure?

Due to the process of cultural conditioning the defined
roles of males and females probably have their psychological
counterparts in the personality organization of individuals
and consequently should be reflected in their self-structures.

3. Is the college experience associated with changes
in the self-structure?

Since the college experience is usually coincident
with the period extending from late adolescence through
young adulthood, and since the college years are in marked
contrast with pre-college life, the resultant change in
personality should be manifest in the self-structure.

A. Is there any correlation between education and

the tendency to regard oneself as an abstraction
rather than as a physical object?

Murphy (15, p. 521) enumerated several hypotheses
relating to the self, and this question was derived from
one of them. Since this eXperimental technique might be
apprOpriate to a question of this sort, it was hOped that
the results of the present study would shed light on Murphy's

hypothesis.

TECHNIQUE AND METHOD

According to the results of two investigators (S, 8),
it was suggested that items representative of the self-
referents relating to the physical self and its appurten-
ances, possessions, group identifications, intellectual
symbols, and affective experiences might be utilized in the
construction of a self-rating schedule. These studies
indicated that items grouped under pertinent categories
tend to have high split-half reliabilities. One study
showed that this technique might reflect both overall
modifications due to living in a collegiate environment
and the effects of other cultural variables.

The schedule was composed of an items which were
assumed to possess face validity. It was hOped that a
structure would be evidenced, showing itself by virtue
of internal consistency within each category and non-
significant or relatively lower intercorrelations between
categories. The definitions of the four categories and
the items comprising each category, numbered as they are
in the schedule, appear below.

1. Intellectual

These items referred to things which are an inte-
gral part of a college student's career and are
concretely related to academic pursuits. One
item, "My belief in free Speech", was indicative
of a broader cultural value.

23.
35.

10.
21.
Al.

13.
31-

29.
26.

20.
an.

l6.
19.
1h.

2. Physical

15

These items referred primarily to the observable
person and his appurtenances.

30 Home

 

These items referred to the home and high mhool

surrounds.

u. C0116g6

These items referred specifically to Michigan State

College and its symbolic representations.

One item,

"College students in general", had a more abstract

referent.

Items Comprising Each Category

 

Intellectual

28.
8.
32.

15.
38.

My grade point average
My present educational
plans

My IQ

My notebook

My belief in free Speech

Physical

My fingernails 18.
My reflection in a mirror 39.
My nose 2h.
The color of my hair 11.
My complexion 36.

Home
My bedroom at home 33.
My high school's 30.
newSpaper 5.
My hometown
My family 90
My old high school 37.

My reputation on campus as
aw

My command of the English
language

A term paper in my major
sub ect

My brains"

The textbook in my favorite
course

A photOgraph of myself

My eyebrows

The clothes I am wearing now
My height

My appearance in a bathing
suit

My high school teachers

My high school's senior prom
The neighborhood I played

in as a child

My high school's football team
My old high school chums

16

Group

17. Michigan State College 22. College students in general
3h. MSC's football team 25. MSC's library
27. The reputation of MSC ho. The name of Michigan State

in the Big Ten College when it appears in
#3. The officers of my class a national magazine or on
7. My immediate circle of the radio

friends 6. President Hannah's success
12. Beaumont Tower as Assistant Secretary of

Defense

Anchoring Items

1. The blood in my veins 2. Blood Spilled on the floor
from a cut in my hand
3. My hair h. Hair on the floor which has

just been cut from my head

The first four items were intended to anchor the
individual's frame of reference and were not scored. All
items were randomly arranged.

The schedule was administered to 129 students attending
Michigan State College. There were 32 subjects each in the
freshman, sOphomore, and junior classes, and 33 in the
senior class. Approximately three-fourths of the schedules
were given to students enrolled in courses taught by the
psychology department of Michigan State College, while the
remainder was solicited from the college dormitories.
Seventy-seven subjects were males and fifty-two were females.

The subjects were instructed to rate each item on a ‘
six-point scale with intervals of twenty percentage points
extending from 0 percent to 100 percent. The six points
were: 0, 20, no, 60, 80, and 100 percent. It was felt that

a scale employing the idea of percentage is more meaningful

17

to the student than are the simple integer scales. The
lowest extreme was defined as representing something that
has nothing to do with the individual, like the moon, while
the uppermost extreme was defined as being a very important
part of the self like the central nervous system. A
diagram was incorporated on the item sheet to guide the

subject's ratings.

RESULTS

As stated previously, certain statistical consider-
ations must be satisfied before this technique can be used
in scientific research. But the resultant statistics are
mainly the empirical material by which the soundness of a
theory may be evaluated. The following discussion will
therefore proceed at the most parsimonious descriptive
level analyzing the data as ratings of items and at an
inferential level interpreting the data as to their

theoretical import.

Question 1.

Are there distinguishable areas of the self-structure differ-
ing in psychological significance to the individual?

If it is assumed, as theory has suggested, that the
self-referents are organized into regions that exhibit a
definable structure, the statistics will reveal the nature
of this structure. Tables I and II indicate that the
ratings of the items organized themselves into a fairly
definite pattern. The relatively high reliability coeffi—
cients of Table I indicate homogeneity or internal consis-
tency within each category. The relatively lower inter-
correlations of Table II show considerable independence
between categories. That is, ratings of items within a
category tend to be more closely related than ratings

between categories. If the items are divided as they were

19

TABLE I

SPLIT-HALF RELIABILITY COEFFICIENTS FOR EACH CATEGORY
CORRECTED BY THE SPEARMAN-BROWN PROPHECY FORMULA

 

 

 

 

Category ' Reliability coefficient
Intellectual .79
Physical .88
Home .76
Group .82
TABLE II

INTELLECTUAL, PHYSICAL, HOME, AND GROUP CATEGORIES
AND THEIR INTERCORRELATIONS

 

 

 

(N = 129)
P H ' G
I .23 (.27)* .35 (.ES) .u7 (.58)
P .15 (.18) .22 (.26)
H 05“ (.68)

 

“Parentheses indicate correction for attenuation.

.17 significant at the 5 percent level of confidence
.22 significant at the 1 percent level of confidence

20

when the reliabilities for each category were computed,
the Split-half reliability for the entire schedule is .90.

Table II shows that when the intercorrelations between
the categories are corrected for attenuation, the consequent
coefficients yield theoretical values indicating the degree
of overlap between the categories. These indices show that
when the sources for unreliability of measurements between
categories are ruled out, the corrected coefficients yield
higher values. Since the reliability coefficients derived
from internal consistency methods are usually underestimated,
the reported coefficients corrected for attenuation probably
are overestimates of the true values. Table II therefore
shows slightly overestimated values of the degree of overlap
between the categories. If the categories overlapped
completely, the coefficient would be one.

Keeping these things in mind, it is clear that the
ratings of items in the Physical category are differen-
tiated from all other categories. In contrast the the-
oretical correlation between ratings of items in the Home
and Group categories is .68, indicating relatively high
overlap. The corrected coefficient between the Intellectual
and Group categories is .58 which is relatively high, and
the coefficient between the Intellectual and Home categories
is a moderately high coefficient of .uS.

On the whole the emergent pattern seems fairly well

established, but the higher relationships need further

elaboration. It was expected that the Home and Group
categories would be completely independent since it was
thought that they tap functionally different areas of the
self-structure, but the data revealed a theoretical corre-
lation of .68. In Spite of this finding there is not com-
plete overlapping, and later results will make this point
clear. Also the overlap between the Intellectual and
Group categories is .58, but since the Group category
contains referents associated with the intellectual atmos-
phere of a college, it is probable that group identifi-
cations become confounded with intellectual referents.
There is little doubt, though, that the Physical category
is a well differentiated area within the self-structure.

Table III gives the mean ratings for each category
and Table IV reveals that the differences between all
categories are significant with the exception of the differ-
ence between the Intellectual and Physical categories. On
a hierarchical scale the Intellectual and Physical cate-
gories rank uppermost followed by the Home and Group cate-
gories in that order. The interpretation pertinent to the
present study is that the regions of the self-structure
have varying degrees of psychOIOgical importance to the
individual.

The significant intercorrelations of Table II should
be considered since no correlation was desirable as indi-

cants of structure. This phenomenon might be regarded from

22

TABLE III
MEAN RATINGS FOR EACH CATEGORY

 

 

 

 

Category Mean
Intellectual 63.5
Physical 60.1
Home h2.0
Group 32.h

TABLE IV

't' RATIOS BETWEEN THE MEANS OF THE INTELLECTUAL,
PHYSICAL, HOME, AND GROUP CATEGORIES

 

 

 

(N = 129)
I P H G
I 1.6 1A.8 23.0
P 8.6 13.6
H 7.A

 

1.98 significant at the 5 percent level of confidence
2.62 significant at the 1 percent level of confidence

23

two viewpoints. First, these data might be attributable

to behavior unique to the schedule. That is, a schedule
frame of reference might have been adOpted encompassing

the upper, middle, or lower parts of the rating scale. Thus
ratings varied within this range irreSpective of categories.
The explanation in accord with this thesis, however, is

that individuals tend to vary in the extent of their self-
involvement -- a psychological condition that cuts across
all areas of the self-structure.

Although the reliability coefficients of Table I with
the possible exception of the Physical category are slightly
below the desired values, it can be assumed that if compar-
ability of items were improved, the coefficients could be
raised to an acceptable level. Intercorrelations of every
item with each other or correlations of each item with the
total category score would indicate those items homogeneous
with one another. If the poor items were eliminated, and
if items with equal intercorrelations as compared to the
other items were added, the addition of about ten such
items to each category would raise all category reliabil-
ities to very satisfactory values. Ideally, the method of
factor analysis should be applied in refining this self-

rating schedule.

Question 2.
Do the sexes exhibit differences in their self-structure?

According to the data of Table V, the differences

TABLE V

MEAN RATINGS FOR EACH CATEGORY BY SEX

21+

 

 

Mean ratings

 

 

Sex
Intellectual Physical Home Group
Male 611.7 57.5 11.2.5 33.3
Female 62.3 63.9 Al.h 31.2

 

25

between sexes for the Intellectual, Home, and Group cate-
gories are not significant. But when a one-tailed 't' test
was applied to the difference in the Physical category, the
difference of 6.h points was Significant at the 5 percent
level of confidence. Here a 't' ratio of 1.66 is required
for significance, and the obtained 't' ratio was 1.66.
Theoretically these results probably indicate that the
Intellectual, Home, and Group categories do not tap those
areas of the self-structure which are sex-linked, while the
differential results in the Physical category suggest that
this area has distinguishable psyChological importance to
each sex. In light of the culturally defined role of women
as passive sought-after objects by the ascendant male, and
considering the culturally ingrained value of physical
attractiveness in heterosexual activities, it seems likely
that the physical referents comprise a sex-linked area of
the self-structure, and the higher value placed upon them
by women is to be expected. There are probably other sex-
linked areas of the self-structure which this schedule has

not reached.

Question_3.

IS the college experience associated with changes in the
self-structure?

The following results were not based upon an ideal
representative sample of the college pOpulation since many

of the subjects were drawn from courses catering to students

26

with Special needs and interests. In regard to this, as
Figure 1 indicates, there is a peak in the Intellectual,
Physical, and Group curves at the junior level. If this
trend is a function of some factor within the non-repre-
sentative sample, the eXplanation for this trend is not
evident in the data. Therefore, since it is dubious to
assume that some selective variable was confounding the
results, the data were interpreted as if they were really
the manifestations of the social psychological processes.

Another source for concern is the possible inadequacy
of controls. Since this study attempted to investigate the
self-rating schedule 223 is, no control group consisting of
a non-college pOpulation was used. This methodological
technique would have indicated whether the college eXperi-
ence was actually the crucial variable Operating. Therefore,
the generalizations derived from the data must, in this
case, be regarded with caution.

AS a partial control, though, the mean ratings for all
categories were computed for age levels. This was done in
order to check whether the impact of the college eXperience
rather than age was the more important variable affecting
the results. The data of Table VI reveal trends presumably
stemming from the college experience while Table VII gives
the aforementioned control data. Figures 1 and 2 are graphs
representing the data of Tables VI and VII, reSpectively.

As a difference of approximately seven points in the

TABLE VI

2?

MEAN RATINGS FOR EACH CATEGORY BY CLASS LEVELS

 

Mean ratings

 

 

 

Class
Intellectual Physical Home Group
Freshman 6102 5303 u308 2803
SOphomore 62.0 62.3 h2.8 33.3
Junior 6801.), 66)..“- (4-306 3803
Senior 62.h 60.3 38.1 29.9
TABLE VII

MEAN RATINGS FOR EACH CATEGORY BY AGE LEVELS

 

 

 

 

Age Mean ratings

level Intellectual Physical Home Group
18-19 62.1 58.2 E6.u 32.2
20-21 6ho2 65.0 h0.0 3h.l
22-26 6h.8 53.6 38.8 28.8

 

Mean ratings

28

 

Intellectual
Physical .___.___.___
Home ______
Group __ ._ .... ..
70 ~—
65 - _/,.\
\

 

55 — /
5O __
45 -
o~ ~~~~~ q _____ ‘\\

35 r- ,/ ‘\

 

l I [I
F s a s

Class levels'

 

Figure 1. Mean ratings for each category by
class levels.

Mean ratings

0’
03

“ J! (n (a
a can 0 018

30

29

 

 

 

 

Intellectual ,__
"—- Physical __ .... ....
Home -----
,_ Group __ - __ -
L. ‘K ——o
/\
/
/ \
- \
s.
P— ‘~\\
__ \\“~~‘
~“~~e
.— -—"""—'—. ’\
0"” ‘\
F'— ‘\‘\.
l l 4
I8 --l9 20 —2l 22-26
Age levels

Figure 2. Mean ratings for each category by age

levels.

30

Intellectual category is required for significance at the

5 percent level of confidence, the upward trend from the
freshman to junior years evidences a significant increased
emphasis on the referents consistuting this area. The
negligible differences in the Intellectual category between
age levels of Table VII probably Show that the college
eXperience rather than age contributes to the increased
ratings of the intellectual referents between the freshman
and junior groups. The decrement of six points between

the junior and senior groups might be due to chance, but
theoretically considered, the intellectual area of the
self-structure probably increases in psychological import-
ance to the college student as a result of his college
eXperience. Then, with the proSpect of graduation and its
practical vocational implications, he "turns away" from his
ivy tower and looks toward the future. The self-structure
is modified accordingly.

A comparison of the mean ratings for each class and
age level in the Physical category indicates a curvilinear
tendency in both cases. Figures 1 and 2 Show this trend
clearly. When the curves are analyzed individually, it
appears that the difference of 11.1 points between the
freshman and junior groups is significant beyond the 5 per-
cent level of confidence, and the difference of l2.h between
the 20-21 and 22-26 age levels is significant beyond the

5 percent level of confidence. In the former case the

31

difference is significant in an incremental direction
whereas in the latter instance, the difference is signifi-
cant in a decremental direction. Since significance is
primarily a matter of degree, the curvilinear tendency
viewed from a molar point of view might give a better
description of the changes in the self-structure than an
atomistic analysis of the situation. In this case parsimony
permits only the statement that there is probably some
determinant modifying the importance of the physical referents
to the individual as he passes through his college career.
Any social-cultural eXplanation as was invoked when con-
sidering the differences in the Intellectual category is
somewhat obscure here.

Both Tables VI and VII and the derived Figures 1 and 2
show a downward trend of mean ratings in the Home category,
but the only difference that is significant is between the
18-19 and 22-26 age levels. About a seven point difference
is necessary here for significance at the 5 percent level
of confidence. From these data it seems that age rather
than the college eXperience is the factor determining the
trend toward lowered mean ratings in the Home category. These
empirical facts probably represent a decline in the psycho-
logical significance that the home and its surrounds have
to the self-structure of the college student. Since a control
group was not used, it is difficult to say definitely whether

this phenomenon is the result of the college eXperience or,

32

as these data tend to indicate, due to age. Further, these
results tend to confirm the notion that with social matur-
ation there is a process Operating which increases the
psychological distance from the home, a condition which
manifests itself in the self-structure.

If this weaning process is axmlid assumption, it might
be eXpected that these former allegiances will be diSplaced
on to the group which has replaced the home environment.

The ascending mean ratings of Table VI in the Group category
tend in part to confirm this eXpectation. The increase of
ten points between the freshman and junior groups is signifi-
cant beyond the 5 percent level of confidence since only
about seven points are needed for significance. Theoreti-
cally this trend indicates that the self-structure assoc-
iated with the college group gains in psychological meaning
to the individual up to a point. Then, interestingly

enough, there is a marked drOp in mean ratings between the
junior and senior groups which is significant beyond the

5 percent level of confidence; and similarly, as Table VII
shows, there is a non-significant decline of 3.2 points
between the 18-19 and 22-26 age levels. A comparable pattern
was evidenced when the mean ratings on the Intellectual
category were analyzed. Here again the college eXperience
rather than age seems to be the crucial variable. The eXplan-

ation for this declining effect probably is that the seniors

33

become concerned with their future plans and the signifi-
cance of the self-referents encompassed by living in a
college milieu is reduced and supplanted by referents pro-
jected into the future. Again the area of the self-structure
and its group referents are modified in the direction of

decreasing importance to the individual.

Question_g.

Is there any correlation between education and the tendency
to regard oneself as an abstraction rather than as a
physical object?

The method and technique of this study are less appro-
priate to this question than they are to the preceding
issues. A preper study would cover a much greater educa-
tional Span, and the abstraction referents should probably
have different content. But the effect might be revealed
by the mean ratings in the Physical and Group categories of
Table VI. Murphy's original hypothesis (15, p. 521) suggested
that the tendency to regard oneself as an abstraction
increases while the tendency to regard oneself as a physical
object decreases with education. The mean ratings in the
Intellectual and Group categories of Table VI show a positive
correlation rather than the hypothesized inverse relation-
ship. There is a significant upward swing between the
freshman and junior levels in the Group category and a
corresponding upward swing in the Physical category between

the freshman and junior groups which approaches significance.

34

Then both means decrease between the junior and senior
groups but only the Group decrement between the junior and
senior levels is significant. If anything, these results
point to a positive correlation between the trends.
Theoretically the Intellectual and Group categories
correlated .58, and, as was pointed out, this was probably
due to the close association between the two in the college
milieu. As Figures 1 and 2 show, both the Intellectual
and Group curves are practically identical. It seems, then,
that some independent variable, possibly the college eXper-
ience, is affecting these areas in a like manner. The Home
and Group categories, which are correlated .68, are probably
not affected by the independent variable in the same way
because, as Figures 1 and 2 indicate, the Home and Group
curves are dissimilar especially between the freshman and ~
SOphomore and the 18-19 and 20—21 age groups. Since the
trends are in the eXpected direction for the Home and Group
categories, their validity as measures of different self-
areas is partially confirmed. The fact that the Physical
category Which was theoretically the most differentiated
area of the self—structure is also similar to the Intel-
lectual and Group curves suggests that the independent var-
iable affects the Physical area similarly. Since the results
of Tables VI and VII and Figures 1 and 2 are based upon
groupings of four and three respectively, any difference
between the curves representing the same category might be

F)

ascribable solely to their grouping.

SUMMARY AND CONCLUSIONS

The purpose of this study was to investigate the use-
fulness of a self-rating schedule as a technique in describ-
ing the self. The need for such a technique was discussed
and the general theory underlying it was examined. A second-
ary intent of the study was to partially answer some questions
derived from theoretical considerations which could be
clarified by this eXperimental schedule. The schedule was
administered to 129 students attending Michigan State College.

The questions posed and the related findings were:

Question 1.

Are there distinguishable areas of the self-structure
differing in psychological significance to the individual?

In general it was found that the subjects' ratings were
organized with regard to four categories of items. These
categories were conceived as representing definable areas
within the self-structure. The four categories were:

1) Intellectual, 2) Physical, 3) Home, and h) Group. The
Split-half reliability coefficients for each category were
.79, .88, .76, and .82 reSpectively. Intercorrelations
between categories were computed, and these values ranged
from .22 to .5h. The relatively low intercorrelations
between categories as compared to the high Split-half reli-

abilities were offered as empirical evidence of an assumed

36

self-structure. The hierarchical order among the four
categories was deemed indicative of the psychological
importance that these areas bear to the individual.
Theoretical problems relating to the independence of the
self-areas and improvement of reliability were discussed.
Since insignificant intercorrelations between categories
were desirable for this study, possible eXplanations for

the significant intercorrelations were advanced.

Question 2.
Do the sexes exhibit differences in their self-structure?

The sample was classified on the basis of sex and
comparisons of mean ratings for each category were made.
The Intellectual, Home, and Group categories revealed
negligible differences between the sexes while the difference
in the Physical category was significant, favoring as might
be eXpected, the female sex. It was surmised that the three
non-discriminating categories are not sensitive to sex-
linked areas of the self-structure, but the Physical cate-

gory is.

Qgestion 3.

Is the college experience associated with changes in the
self-structure?

The cross-sectional approach was applied, utilizing
the four class levels as representative of a develOpmental

sequence. A partial check was employed whereby differences

37

related to age were compared to class level differences.

The possibility of bias in the sample was noted and due
caution was advised. Developmental trends were considered
for each category, and significant differences, when they
appeared, were discussed. Interpretations and eXplanations
were advanced for the empirical phenomena. The decrement
between the junior and senior classes and the 20-21 and
22-26 age groups necessitated a specific interpretation.

In general it was suggested that the Intellectual area of
the self-structure increases in psychological importance

to the college student until the junior year when a decrease
occurs. A similar curvilinear trend was found for the
Physical area. The Home area appeared to decrease in
psychological significance to the individual with time.
DevelOpmentally the group area of the self-structure appeared
to correlate with the intellectual area in its psychological
significance to the individual.

Question h.l

 

Is there any correlation between education and the tendency
to regard oneself as an abstraction rather than as a
physical object?

Although this technique is not ideally apprOpriate to
a question framed in this way, the mean ratings between the
Physical and Group categories appear to be positively corre-

lated for succeeding class and age levels rather than inversely

related as was implied by Murphy.

38

It may be concluded that with refinement this eXperi-
mental self-rating schedule can be made sufficiently reli-
able to warrant its use. Its validity, as an instrument
in revealing the self-structure, may then be investigated.
Some results of this study tended to corroborate what theory
might have indicated, and this partially, at least, attests
to its validity. The main advantage of this schedule is-
its ease of administration and simplicity, but, on the
other hand, it is limited because it is useful only with a
college pOpulation. Further research with other populations

and behavioral validation in experimentally controlled

situations are indicated.

l.

2.

3.

h.
S.

6.

7.

8.

9.

10.

11.

12.

13.

1h.

 

REFERENCES

Allport, G. w. Personality: 5. s choloeical inter-
pretation. New York: Henry HoIE, I957.

 

____ . The ego in contemporary psychology.

PSYChOle 59-2., 1911.3, 50, uSI‘LL76.

Brownfain, J. J. Stability of the self-concept as a
dimension of personality. (Abstract). Microfilm

 

 

Calkins, H. w. 5 first textbook in psychology. Ed. A.
New York: Macmillan Company,m19IE}

 

 

Gladin, L. L. Unpublished manuscript. Michigan State
College, East Lansing, 1952.

Grossman, D. The construction and validation ofStwo
insight inventories. ‘J. consult. Psychol., 19 l,
15, 110-115.

Guilford, J. P. Fundamental statistics 32 s cholo
and education. Ed. 2. New York: McGraw-HEll, 195%.

 

Hartocollis, P. Unpublished manuscript. Michigan
State College, East Lansing, 1950.

Hilgard, E. R. Human motives and the concept of the
self. Amer. Psychol., 19h9, h, 37h-382.

Holt, R. H. The accuracy of self-evaluations: Its
measurement and some of its personological corre-
lates. ‘J. consult. Psychol., 1951, 15, 95-101.

James, W. Principles pf psychology. Vol. 1. New York:
Henry Holt, 1690.

 

 

Koffka, K. Principles of Gestalt psychology. New York:
Harcourt, Brace, 1935.

 

 

Lewin, K. D amic theo [pf personality. New York:
McGraw-HI¥§, 193 .

. Principles of tgpological psychology. New
York: McGraw-Hill, 1936.

15.

16.

17.

18.

19.

20.

21.

22.

23.

ho

Murphy, G. {A biosocial approach pp origins and
structures. New York and London: Harper and Brothers,

I947.

Philips, L. E. Attitudes toward self and others: A
brief questionnaire report. J. consult. Psychol.,

19511 15: 79'810

 

 

 

 

Plant, J. S. Personalipy and the cultural pattern.
New York: Commonwealth Fund, 1937.

 

Rogers, 0. R. Client-centered psychotherapy. Sci.
Amer., 1952, 187, 66-711.

Sheerer, E. T. The self concept and feelings toward
others. .l’ consult. Psychol., 1949, 13, 169-175.

Sherif, M. and Cantril, H. The psychology 2; ego-
involvement. New York: John Wiley and Sons, 19h7.

 

Snygg, D. and Combs, A. w. Individual behavior: A
new frame of reference. New York: Harper and
Brothers, I9h9.

Stagner, R. Ps cholo pf personality. New York:
McGraw-Hill, E9KB, r

 

 

 

 

Stock, D. The self concept and feelings toward
others. ‘1. consult. PsyChol., 19h9, 13, 176-180.

APPENDIX

EXperimental Schedule

This is a study of your SELF. You are asked to report what things. objects.
and groups are considered part of yourself and.what you consider foreign to
' you. The scale of closeness to You runs from 100% for something which is en-
tirely part of you, like your nervous system, to 0% for something that has
nothing to do with you, like the moon. Use the six percentages as shown in
the diagram: 10033, 80%, 60%, 40%, 20%. 0%. Example: "My shoes" If you think <
of your shoes as part of you, as being very close to the real you, rate them
high, perhaps 60% or 80%. If you feel that your shoes have very little to do
with you, and could just as well be part of someone else, rate them low, perhaps
20% or 40%. The question is pg; how much.you.like these things nor how good
they are but how intimately these things are included in.your concept of your
self. Would you be a different person if these things were different? Rate all
items.

1"“

Fill in identifying data on top of test sheet.

 

M _ f
‘

 

 

LL

 

 

 

 

 

 

 

 

’“049~W“_.x-_. ...i. . m -.n...i:-,,, ., ,__

 

 

 

 

 

0000214..
....25.
....26.
....27.
0.00280
....29.
000030.
....31.
....32.
000033.
....34‘.
0000350
0000360
....37.
....38.
e00039o

_‘ ....40.

....41.
000042.
0.0043.

.....MO

Class Major:

 

Age:

Sex:

 

The blood in my veins

Blood spilled on the floor from a cut in my hand.

My hair

Hair on the floor which has just been cut from my head.

The neighborhood I played in as a child

President Hannah's success as Assistant Secretary of Defense
My immediate circle of friends in college

My command of the English language

My high school's football team ,r":fl:-2° ‘KR‘ \.
my IQ ,W/ 45‘ ‘s. ‘\
My height ‘ ~- e, x ‘

Beaumont Tower

My fingernails

My old high school

My "brains"

My hometown

Michigan State College
A.photograph of myself
My family

My bedroom at home

My notebook

College students in general
My grade point average

0

 

The question is not how good these things are but how intimately these
things are included in your concept of your self.

The clothes I am wearing now

MSC's library

My complexion

The reputation of M80 in the Big Ten

My reputation on the campus as a student
The color of my hair

My high school's senior prom

My reflection in a mirror .

vA term paper in my major subject

My high school's teachers

MSC's football team

My present educational plans

My appearance in a bathing suit

My old high school chums

The textbook in my favorite course.

My eyebrows

The name of Michigan State College when it appears in a national magazine
or on the radio

My belief in free speech
My nose

The officers of my class
My high school's newspaper

 

 

 

 

 

 

 

 

 

 

Mme» 5m: 5

J:XZC:»\JA1_S_- "DP-”.m—
' ‘ {amber _ z. ‘

Us 3 '52
, as: 1 More;

~55 Aou Ze—Jtmwfaa

AP 24

 

- I ~ 2’35
“g. ' “t" . ., ..

Mar "‘ 5‘“

   

APR 2853)
Ear 10 58

 

 

 

 

 

 

I
III
I
l
III-III
I
ll
I

II

III

177 9444

 

"Minimum