SOME PSYCHOLOGICAL AND PHYSIOLOGICAL FACTORS
AFFECTING EXCELLENCE IN ACTING

Thesis for the Degree of Ph. D.
MICHIGAN STATE UNIVERSITY
WILLIAM H. STOCK
1968

$st

.— I

 

IIIIIIIIIIIIIIIIIIIIIIIIII\IIIIII 3;; ~* I

f.

 

This is to certifg that the
thesis entitled

SOME PSYCHOLOGICAL AND PHYSIOLOGICAL
FACTORS AFFECTING ACTING

presented by

William H. Stock

has been accepted towards fulfillment
of the requirements for

PHD degree in Theatre
Date 6122168

 

0-169

 

 

 

 

 

 

 

 

 

 

 

7‘ " ; '3 *1"?!

(~qu 3933'"

 

ABSTRACT

SOME PSYCHOLOGICAL AND PHYSIOLOGICAL FACTORS
AFFECTING EXCELLENCE IN ACTING

by William H. Stock

This study was predicated on the assumption that
there do exist certain measurable differences between
actors and non-actors, and also between persons with
high and low potential in acting. The purpose of this
study was to examine a group of psychological and physio-
logical factors within a controlled observational situ-
ation in order to determine the extent to which these
factors constitute measurable differences.

Thirty—five factors ranging from ability to
accurately differentiate differences in pitch to psycho-
logical flexibility were examined. Data were gathered

with six separate testing instruments; The Seashore

 

Measures of Musical Talents, The Otis Quick-Scoring

 

Mental Ability Tests, Gamma Fm, The MacQuarrie Test for

 

 

 

Mechanical Ability, The Remote Associates Test, The

California Psychological Inventory, and The Stock Index

 

 

of Solipsism. Each separate sub-division or scale of

 

each test was regarded as a separate factor.

William H. Stock

Data were gathered from a sample of 100 persons.
The sample was divided intdxfour observational groups of
twenty-five persons each. The observational groups were
categorized as actors of proven worth and ability, be-
ginning actors with high potential, beginning actors with
low potential, and non—actors. The actors of proven
worth and ability were drawn from the Performing Arts
Company of Michigan State University and the Trinity
Square Repertory Company of Providence, Rhode Island.
The beginning actors of high and low potential were drawn
from undergraduate students of acting at Michigan State
University. They were placed in either the high or low
category on the opinion of faculty Judges. The non-
actors were drawn from the general student population of
Michigan State University. Members of this observational
group were individually questioned to determine that they
had never acted, and that they had no desire to do so.

The six testing instruments were administered to
the sample in several testing sessions. The data obtained
were analyzed by analysis of variance to determine what
factors constituted significant differences. Criterion
level was held at alpha .05.

Four factors differentiated significantly between
actors and non—actors. They were ability to differenti-
ate rhythms, sense of responsibility, communality, and

solipsism. Actors were better able to distinguish

William H. Stock

rhythms, had a lower sense of responsibility, a lower
feeling of communality, and less tendency toward solipsism.
It was concluded that these four factors constituted a
measurable difference between actors and non-actors.
Neither these factors nor the other factors under con-
sideration successfully differentiated high from low
potential actors.

An additional four factors differentiated between
actors and non-actors, but only at the 10 per cent level.
These factors were Tonal Memory, Self-acceptance, Sociali-
zation, and Feminine Interest. It was concluded that
further research was needed in these factors to determine
the extent to which they constitute measurable differences

between actors and non—actors.

SOME PSYCHOLOGICAL AND PHYSIOLOGICAL FACTORS

AFFECTING EXCELLENCE IN ACTING

By

William H. Stock W

A THESIS

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

DOCTOR OF PHILOSOPHY

Department of Speech and Theatre

1968

ACKNOWLEDGMENTS

To Dr. E. C. Reynolds for his understanding
guidance

To Dr. James McCroskey for patience in
helping me to understand

To my wife for putting up with it all

This study is dedicated with my deepest respect and

gratitude.

ii

TABLE OF CONTENTS

Page
ACKNOWLEDGMENTS . . . . . . . . . . . ii
LIST OF TABLES . . . . . . . . . . . iv
Chapter

I. INTRODUCTION. 1
Purpose . . . . . . . . 1
Limitations . . . . . . 1
Importance of the Study . . . . . 3
Definitions . . . . . . . . . 6
Organization. . . . . . . . 9
II. REVIEW OF THE LITERATURE. . . . . . 10
Literature in Acting . . . . . . 10
Literature in Psychology. . . . . 16

III. SUBJECTS, INSTRUMENTATION, AND
PROCEDURES . . . . . . . . . 32
Subjects . . . . . . . . . . 32
Instrumentation. . . . . . 38
Procedures . . . . . . . . . 66
IV. RESULTS . . . . . . . . . . . 72
V. CONCLUSIONS . . . . . . . . . . 81
BIBLIOGRAPHY . . . . . . . . . . . . 97
APPENDIX . . . . . . . . . . . . . 105

iii

Table

10.

ll.
l2.
13.
IA.
15.
16.

LIST OF TABLES

Values for t for the Individual Factors
Tested . . . . . . . . .

Values for U for the Individual Factors
Tested

Frequency Differences Between the Pairs--
Test for Pitch . . . . . . .

Intensity Differences Between the Pairs--
Test for Loudness . . . .

Rhythmic Patterns--Test for Rhythm.

Differences in Duration Between the Pairs
-—Test for Time . . . . . . .

Decibel Variations from the Levels in the
Standard Tone Used in the Variable
Tones-—Test for Timbre . . . . .

Consistency Coefficients for College
Level Students. . . . . . , . .

Reliabilities Determined by MacQuarrie by
the Test- retest Method, Based on 365
Cases. . . . . . . . . .

Correlations Based on One Study with
200 Subjects . . . . . . .

Parametric Analyses of Variance.

Results of tftests . . . . . .
Non-parametric Analyses of Variance

Rank Sums . . . . . . . . . .
Results of U Tests . . . . . . . .

Results of U or tgtests, Group 1

iv

Page

3A

36

Al

Al
A2

“3

“3

45

SA

61
75
76
77
78
79
80

CHAPTER I

INTRODUCTION

Purpose
This study is predicated on the assumption that

there do exist certain measurable psychological and
physiological differences between actors and non-actors,
and also between persons with high and low potentials
for success in acting. The purpose of this study is to
examine a group of psychological and physiological
factors within a controlled observational situation in
order to determine the extent to which these factors

constitute measurable differences.

Limitations

 

This study is an attempt to establish certain
measurable differences in a creative field where little
previous quantitative study has been carried out. The
problem is a complex one. The infinite number of possible
factors involved in the art of acting seemingly precludes
that a single study could encompass all possible factors.

There exist at least two distinct approaches to

the problem. One is the approach of this study, which

attempts to use existing group testing instruments to
determine the existence of differentiating factors in
groups selected according to the level of their skill
and potential in acting. The other approach, which in
the case of some physiological factors might be more
fruitful, would be to use medico-physiological machines
like the Technicron Multiple Sequential Blood Analyzer
to seek differentiating factors. The present study is
limited to the first approach for several reasons.

First, while the medico-physiological approach may
be more valid for certain physiological factors, it can-
not by its very nature reveal the existence of factors
within the equally important psychological area, which
is the main focus of this investigation. Second, both
the use of such machines and the time of technicians who
are needed to operate them are prohibitively expensive
for the researcher working without benefit of funding.
Finally, reviewing the literature which does exist has
led this researcher to conclude that there is a strong
possibility that there may be more detectable factors in
the psychological area than in the medico-physiological
area.

The single exception to the use of existing re-
search instruments in this study is the Stock Index of

Solipsism. The SIS is a psychological testing instrument

 

which the researcher deveIOped specifically for this

study. It was develOped both because there was no other
instrument available in the factor area, and to give the
researcher experience in test development. The instru—

ment and its development are discussed in Chapter III.

Importance of the Study

 

That there has been a sharp increase in theatrical

activity in recent years both in the live and mechanical
media is a fact of which even those not directly concerned
with the theatre are aware. Because of this increased
interest, attention has been given on several fronts to
serious deficiencies in the way in which we as a nation
train young artists who desire a place in the world of
the professional performing arts. Perhaps the most notable
of these is the recent report of the Rockefeller Panel on
the Performing Arts. The Panel, which was composed of
leaders in all of the performing arts, made the following
comment on the present status of performing arts education:
If the performing arts are to fulfill their cultural
mission in the United States, marked improvement in
the quality of the training of professional artists
will be required. It has been authoritatively
asserted that much of the dance instruction avail-
able in this country is harmful physically as well
as aesthetically. In the theatre there is wide-
spread complaint of ill-trained craftsmenship on
the part of those seeking professional status.1

Despite this rather dark View of the present state

of performing arts education, the Panel outlined several

 

1The Performing Arts, Problems and Prospects (New
York: McGraw—Hill Book Company, 1965), p. 18.

 

hopeful prospects for the future. Among these prospects
is the increasing commitment being made to training the
performing artist by American colleges and universities.
The Panel feels these institutions, ". . . have begun
to assume increasing responsibilities for training
future performers."2
Colleges and universities have encountered some
problems in assuming this new responsibility. Despite
the fact that change is taking place, an attitude still
exists in some faculty quarters which is inimical to pro-
fessional training in the performing arts. Such training
is equated with "skills" training, which should be rele-
gated to "trade schools" of the arts. Persons of the
opposite persuasion are hindered in replying to these
charges because at present no objective criteria for
measuring aptitude in the performing arts exist. There
are on the other hand, objective criteria for measuring
aptitude for more traditional approaches to scholarship.
Thus admissions procedures are weighted against the stu-
dent with little aptitude for traditional university study,
but high aptitude for the performing arts.
The difficulty of determining who may be profitably

admitted to a professional training program in the per—

forming arts in general, and particularly in theatre, is

 

2Ibid., p. 173.

not unique to university departments. It is shared by
private professional training schools. Both are faced
with the troublesome task of working out admissions pro-
cedures which are both fair and fruitful in an area
where quantitative measures of potential excellence are
largely non—existent.

The two horns of the theatre educator's dilemma
are manifest in the following statements from the Rocke-
feller Report and the American Educational Theatre
Association's Conference Report: "At present admissions
criteria in the university emphasize academic attainments
and have little relevance to creative ability in the
arts."3 While on the other hand:

Research is needed on the possibility of develop-

ing psychometric instruments which will measure at

some useful level the probability of success in
actor training programs . . . or on the develop-
ment of new instruments capable of identifying
those children who should be encouraged to seek
careers in the arts, or discouraged from pursuing
such career aspirations.
In effect, if one can accept the authority of these two
studies, persons conducting actor training programs are
faced with the fact that existing standards for ad-
missions bear little relationship to identifying indi-

viduals with a potential for success in professional

 

3Ibid., pp. 176—177.

 

“Kenneth L. Graham, "Relationships Between Edu-
cational and Professional Theatre; Actor Training in the
United States," Educational Theatre Journal, Special
Issue (November, 1966), p. 332.

training programs. They are also faced with the fact
that the development of these new admissions procedures
is being delayed by a failure to develop or discover
appropriate measurements of potential for success at
any meaningful level.

Obviously, the first step in developing such a
set of measures is to establish what things may profitably
be measured. In other words, before measuring instruments
may be developed it is necessary to establish what factors
differentiate the successful actor from the unsuccessful
one. It is necessary to determine quantitatively and ob—
jectively what our judgments reflect about the behavior
of the person whom we judge to be capable of excellence in
the art of acting. This study seeks to establish a group
of these factors, and thus to take a first step toward

establishing meaningful measures of acting potential.

Definitions

 

The term "proven actor" will be used to indicate an
actor who has performed in at least twenty roles, who
earns money by the art of acting, and who is found by a
significant majority of the directoral staff of the pro-
ducing organization by which he is employed to be a more
than usually talented actor or actress.

The term "beginning actor with small potential"
will be used to indicate persons who have been judged,

either on the basis of class work, performance, or both,

to be clearly deficient in potential excellence in acting
by the faculty in acting of Michigan State University.

The term "beginning actor with large potential"
will be used to indicate persons who have been judged
on the basis of both class work and actual performance
to be clearly possessed of large potential for excellence
in acting by the faculty in acting at Michigan State
University.

The rationale for the definitions in the two para—
graphs above is that the only real basis which exists at
present for differentiating between high and low potential
in acting——either potential for the successful completion
of a course of study or potential for the successful
rendition of a specific role by an experienced actor-~13
a subjective decision by an experienced producer, director,
or teacher. Since these subjective decisions are the
basis of most current theatrical selection, there seems
little choice but to allow them to form the basis for
the categorizations of subjects in this study.

The term "non—actor" will be used to designate
those persons, drawn from the Michigan State University
student population who have never acted in a play, and
who, moreover express no desire to do so.

"Acting" will be taken to mean the act of creating
roles within plays, film plays, or television plays. It

will not be taken to include the conscious or unconscious

role—playing which is a part of the daily life of most
normal human beings, or the peripheral role-playing of
such performers as night-club comedians or radio
announcers. While it is freely admitted that a penchant
for such activity might bear some relationship to ex-
cellence in acting, the examination of that relationship
is held to be outside the purpose of this study.
"Creativity" will be used as a term in the ex-
.tremely limited sense proposed by Dr. Sarnoff Mednick
of the University of Michigan. Dr. Mednick views
creativity as a process "of seeing relationships be-
tween seemingly 'mutually remote' ideas and forming them
into new associative combinations which are either use-
ful or meet specified criteria."5
The term "factor" will be used to indicate an area
of human behavior in which a measurable difference be-
tween actors and non—actors is hypothesized to exist.
”Physiological factors" will be taken to mean those areas
of human behavior which are primarily governed by the
makeup of the subject's body, as in the ability to hear
sounds of a given pitch level. "Psychological factors"
will be used to indicate those areas of human behavior

which are primarily dependent on the particular bent of

 

5Sarnoff A. Mednick and Martha T. Mednick,
Examiner's Manual Remote Associates Test (Boston:
Houghton Mifflin Company, 19673, p. l.

 

the subject's mental makeup, as, for example, the tendency
to attempt to dominate a social group.
For purposes of this study, all of the above terms

are strictly limited to the above stated definitions.

Organization

 

The study will be divided into five chapters with
appendices and tables. The five chapters include the
present introduction, a review of the literature in the
area of the study; a chapter discussing the subjects,
the testing instruments, and the experimental procedures
used; a chapter devoted to the results of the experiment;
and finally, a chapter of summary, conclusions, and sug-

gestions for further research.

CHAPTER II

REVIEW OF THE LITERATURE

This chapter is a review of the existing literature

in the area of this study. It is divided into two main

sections; literature in acting and literature in psychology.

Literature in Acting

 

In approaching the present study, there is a strong
temptation to disregard entirely the past literature in
acting. In effect, the literature in the field of acting
has primarily concerned itself with questions of method.
It has asked, and has attempted to answer, questions re-
lating to hgw the actor performs the task of moving an
audience. While this question remains a vital one, it is
not the main focus of this inquiry. This study seeks not

to discover what means an actor may best use, but rather

attacks the question of what peculiarities of human consti-

tution contribute significantly to what is called "talent
in acting." It is an inquiry not into the actor's means,
but into the nature of the actor himself. As such, it is
somewhat remote from the question of means so prevalent

in the literature. Naturally, it must be added that the

discovery of quantitative data about the nature of talent

lO

11

may ultimately affect the question of the means by which
that talent may best be employed.

In brief, the traditional literature in acting, in
attempting to answer the question of means, polarized at
the earliest possible date into two general schools of
thought. These two camps, which have used differing
terminologies depending upon time and place, may be
categorized as the inspirational view and the view of
technique. The former holds that the actor best affects
the audience when he is inspired to feel the emotions of
his role. The latter holds that the actor should not feel
the emotions, but rather should make the audience feel
them by the artistic use of his physical instrument.

In the third century before Christ, Plato concluded
in the log that actors, poets, and rhapsodes are directly
inspired from above, and thus may work without the aid of
art. Indeed, this researcher has taken the terms "inspiration"
and "art" from Plato. Plato's view of direct inspiration
was magnificently lampooned by Aristophanes in his charac—

terization of the inspired poet in The Birds, the "zealous
l

 

slave of the Muses." The View implied by Aristophanes'
satire was stated in the following somewhat more philo-
sophic form by Plutarch in his Symposiacs four centuries

later:

 

lAristophanes, The Birds (tr. anonymous), line 904.
In Whitney Oates and Eugene O'Neill, Jr., The Complete
Greek Drama, Vol. 2 (New York: Random House, 1938), p.
768.

 

 

 

l2

. he that only represents excels him that

really feels, inasmuch as he doth not suffer

the misfortunes; which we knowing are pleased

and delighted on that account.2
Thus both points of View were quite firmly established
even in the ancient world.

The pattern of opposing inspiration to art con-
tinues through the literature in the oppositions of
Horace versus Cicero, Kean versus Kimble, Irving versus
Coquelin, down to the Stanislavski versus Dean contro-
versy of recent times. While the contemporary critic of
acting would usually recognize that, taken alone, either
side of the argument represents only half of the truth,
there still exists some suspicion that one way or the
other may be superior.

The question of the actor's means has been the sub-
ject of some excellent contemporary scholarly research,
which has used not only the more traditional critical
methodology, but also the newer means of creative and
quantitative research. Isaiah Sheffer made good use of
the traditional critical methodology in his "Emotional
n3

Memory in Acting Technique. Sheffer, by gathering and

examining the existing literature in emotional memory,

 

2Plutarch, Symposiacs, Book V, cited by Toby Cole
and Helen Chinoy in Actors on Acting (New York: Crown
Publishers, 1949), p. 13.

 

 

3Isaiah Sheffer, "Emotional Memory in Acting
Technique" (unpublished Master's thesis, Michigan State
University, 1958).

l3

managed to clarify the actor's use of this technique,
but, in his own words, "The project of trying to verify
the suppositions of the theatrical artist in the light
of modern psychological knowledge is specifically ex-
cluded."u

"Creative" research into the actor's means may
take the form of an actor-student's describing his
creative methodology in the task of forming a single role,
as in Mariam Alexanian's 1952 study of the role of
"Elizabeth the Queen."5

As early as 1952, Whitehall and Kodman attempted,
with somewhat inconclusive results, to make a quantitative
analysis of the reactions of audiences to stereotype
character.6 Only five years ago, Allen Neal Kepke suc-
cessfully applied quantitative methodology to the problem

of character communication by the use of the Q-sort

technique.7 This interesting study explores the

 

ulbid., p. 3.

5Mariam Alexanian, "Elizabeth the Queen" (unpub-
lished Master's thesis, Michigan State University, 1950).

6Buell Whitehall, Jr., and Francis Kodman, Jr.,

"A Study of Audience Reaction to a Stereotype Character,"
Educational Theatre Journal, Vol. IV, No. 2 (December
1952), 139-192.

 

7Allen Neal Kepke, "A Study of Communication of
Perception of Character Among Actors, Director, and
Audience Using Q Methodology" (unpublished Ph.D. dis-
sertation, Michigan State University, 1963).

1A

difficulties in communicating concepts of character with
enviable thoroughness, but remains within the traditional
literature in confining itself to the study of the actor's
means rather than the actor's nature.

It is only in very recent times that writers on
acting have turned from the pattern of inquiring after
the actor's means to question what an actor is, and what
makes up the "being" of a successful actor. Perhaps the
most notable of these works is Yoti Lane's The Psychology

of the Actor.8 While this work suffers, from the view-

 

point of this researcher, from deficiencies of methodology,
it must be regarded as something of a landmark in the
literature in its effort to describe the psychological
makeup of the actor. Lane's "analysis" of the problem
consists of a series of observations, and deductions from
those observations, set down in a pseudo-psychoanalytic
style.

9 and James Ching,lo which

The works of Jerry Blunt
seek to describe the actor's psychology, suffer from the

same methodological deficiency. Blunt, for example,

 

77

8Yoti Lane, The Psychology of the Actor (New York:
John Day Company, 1960).

 

9Jerry Blunt, The Composite Art of Acting (New
York: Macmillin, 1966), p. 23.

 

10James Ching, Performer and Audience (Oxford:
Hall Publishers, l9A7).

 

15

speaks of the actor's need for "sensitivity," but makes
no attempt to define the nature of that sensitivity, or
that to which the actor need be sensitive. While these
observations have the ring of truth, they provide no
more objective data about the nature of actors than the
considered judgment of any experienced theatre person.
Despite deficiencies of method and lack of ob-
jectivity, the present study is deeply indebted to these
authors, and to Lane especially, for providing invaluable
insights as to which factors to investigate. It is hoped
that this study, and additional quantitative research,
may help to make a meaningful proof of their observations.
Before leaving the literature of the theatre area,
it should be noted that there have been growing efforts to
apply experimental methods to studies in the field. These
studies have been various in scope and application. Both
E. C. Mabie and Raymond Smith attempted to use the reaction
of the audience during performance through the use of re-

sponse indices. Mabie worked in the area of audience,
ll

 

reaction, while Smith attempted to develop a semantic
differential for use in the theatre area.l2 Besides the
11

E. C. Mabie, "The Responses of Theatre Audiences,
Experimental Studies," _peech Monographs, Vol. XIX, No.
A (November 1952), 235—2A3.

12Raymond G. Smith, "A Semantic Differential for
Theatre Concepts," Speech Monographs, Vol. XXVII, No. 1
(March 1961), l—8.

 

16

work of these two men, studies have ranged from inquiries
into the nature of attenders and non-attenders of college
theatres to the author's own modest attempt to develop a

13

new base for scene paint. These studies are mentioned
in passing since they contribute to a growing confidence
in the use of experimental methodology within the field,

although not a part of the literature of acting.

Literature in Psychology
Ten minutes spent reading titles in any university
library will confirm that the literature in the behavioral
sciences, is, to say the least, vast. Catalogue listings
in the psychology of creativity alone range from the

philosophical Creative Intuition in Art and Poetrylu by

 

Jacques Maritain to the somewhat unscholarly How to Be a

 

More Creative Executive15 by Joseph Mason. Since any
brief attempt to survey so huge a sea of information is
doomed by the very size of the area to be charted, this
discussion will be limited to the two areas which are

most relevant to the present study: the literature in

 

V

13William H. Stock and Robert Grubbs, "MLL, A
New Base for Scene Paint," Educational Theatre Journal,
Vol. XVLL, No. A (December 1965), 237-239.

 

l“Jacques Maritain, Creative Intuition in Art and
Poetry (New York: Pantheon Books, 1953).

 

15Joseph Mason, How to Be a More Creative Executive
(New York: McGraw-Hill, 1960).

l7

creativity, and the literatures surrounding the testing
instruments which form the basis for this study.

Within behavioral science research, there is little
unanimity of opinion as to the root definition of
creativity. Not a little of the literature is devoted
to seeking just such a root definition. Since this
study has chosen to adopt one particular definition of
creativity,16 and since, as Alan Downer remarked at the
recent conference on theatre research, "Conversations on
this subject revealed that we were rather sure that many
kinds of creativity were involved in the theatre,"17 a
detailed treatment of the varying opinions as to what
constitutes creativity would seem superfluous. For a
complete, concise, and comprehensive discussion of the
varying points of view on the subject of creativity, it
is suggested that the interested reader consult Morris
Stein's Survey of the PsychologicalfiLiterature in the Area

of Creativity.18

 

Turning to the rationale of the definition of

creativity which this study accepts, the noted acting

 

16See p. 8.

17Alan S. Downer, "Conference on Theatre Research,"
Educational Theatre Journal, Special Issue (June 1967), p. 2A7,

l8Morris I. Stein, Survey of the Psychological
Literature in the Area of Creativity (New York: Mac-
millin, 1962).

 

l8

coach, I. Rapoport, made the following remark on the
creative working of the actor's imagination:

It is impossible to imagine something non-exis-
tent, but to unite different parts of the exist—
ing into a new whole--creating thereby a new
"artistic image"--can be done with the help of
the creative fantasy.l9

This point of view agrees with the definition of crea-
tivity which this study accepts: Sarnoff Mednick's

notion of creativity as facility for bringing together

mutually remote things in new and fruitful ways.2O

Beginning in 1958 with "Ease of Attainment of

Concepts as a Function of Response Dominance Variance,"21

and "An Orientation to Research in Creativity,"22 Dr.

Mednick, working in collaboration with his wife, J. L.

 

191. Rapoport, "The Work of the Actor," in Acting:
A Handbook of the Stanislavsky Method, edited by Toby
Cole (New York: Lear Publishers,'l9A7), p. A2.

 

20The repeated attribution of this definition to
Dr. Mednick is a matter of linguistic facility, and is
by no means meant to underestimate the contributions of
Dr. Mednick's many collaborators who helped in the for—
mation and refining of the concept, among whom J. L.
Freedman and Dr. Martha Mednick should be mentioned as
most notable.

21J. L. Freedman and Sarnoff A. Mednick, "Ease of
Attainment of Concepts as a Function of Response Domi-
nance Variance," Journal of Experimental Psychology,

LV (1958), A63—A66.

 

22Sarnoff A. Mednick, "An Orientation to Research
in Creativity," Research Memo No. 2 (Berkeley, Cali-
fornia: University of California, Institute of Per-
sonality Assessment and Research, 1958).

l9

Freedman, and others,23 began to explore and expand the
"remote associates" definition of creativity. Working
mainly from the idea of scientific creativity, Mednick
and his associates repeatedly tested the validity of
their definition of creativity until they were able to
apply it satisfactorily to a testing instrument which
would seek to measure creativity as they had defined it.

This instrument is the Remote Associates Test which

 

Houghton Mifflin published in 1967, and which this

study is attempting to use with reference to the problem
of acting. For a complete listing of the literature
leading to its development and refinement, the interested
reader is referred to the bibliography of the present
study and to that of the Mednick's study, "The Associative

Basis of the Creative Process."2u

It is to be hoped that
this study will help to make clear the relationship of
creativity, as Mednick defines it, to the art of acting.

The interested reader whose technical competencies
do not include a general knowledge of the area of testing
may find an excellent introduction to the subject in

25

either Anne Anastasi's Psychological Testing, or in

 

23Notably C. C. Jung, S. Halpern, and J. P.
Houston.

2”Sarnoff A. Mednick and Martha T. Mednick, "The
Associative Basis of the Creative Process," HEW Coopera-
tive Research Project No. 1073, University of Michigan,
1965 0

25Anne Anastasi, Psychological Testing (2nd ed.;
New York: Macmillian, 1961).

 

2O

Edgar Anstey's Psychological Tests.26 Either book can

 

provide a good general view of recent thinking in the
field.

The Seashore Measures of Musical Talents is one of
the oldest and most widely used testing instruments still
in print. It was first developed by Carl E. Seashore in
1919, and later revised in 1939 to make the stimuli to
which the testee is asked to respond more precise. A
detailed discussion of the construction, content, and
analysis of the 1939 revision, which is still current,
and which is used in this study, is available through

27

the University of Iowa Press. The Seashore Measures

 

has been repeatedly tested to determine both its relia-

bility and validity, and to determine its usefulness in

various kinds of testing.28

 

26Edgar Anstey, Psychological Tests (London:
Nelson, 1966).

 

27D. Lewis, J. G. Saetveit, and C. E. Seashore,
"Revision of the Seashore Measures of Musical Talents,"
University of Iowa Studies Aims Progress Research No.
65 (Iowa City: University of Iowa Press, 19AO).

28Vide S. E. Farnum, "Prediction of Success in
Instrumental Music" (unpublished Ph.D. dissertation,
Harvard University, 1950); A. S. Kane, "A Correlative
Study of Musical Aptitude and Music Appreciation as
Measured by the Seashore Measures of Musical Talents
and the Fisher Measurements of Musical Appreciation"
(unpublished Master's thesis, Boston University, 1950);
J. E. Karlin, "The Factorial Isolation of the Primary
Auditory Abilities" (unpublished Ph.D. dissertation,
University of Chicago, 19A2); and W. S. Larson,
"Practical Experience with Music Tests," Music Edu-
cation Journal, XXIV (1938), 68-7A.

 

 

21

Normative information for the Seashore Measures

 

has been obtained by testing programs in some nineteen
secondary school systems and colleges.29
Dealing as it does with the ability to hear cer-

tain aural stimuli with precision, the Seashore Measures

 

obviously deals with the most physiological of the

factors examined by this study. The Seashore Measures

 

has been used with success to study problems as diverse

30

as aptitude testing in music, aptitude testing in

32 33

telegraph,31 spelling competence, and aphasia.

 

29They are the Northbrook, 111.; Grand Haven, Mich.;
Spring Lake, Mich.; Onamia, Minn.; Harlowtown, Mont.;
Niagra Falls, N. Y.; Olean, N. Y.; and Windsor, N. Y.
school districts, and the University of Maine, Union
College, and Caspar Junior College.

3OVide May V. Seagoe and J. C. Gowan, "The Relation
Between IHFErest and Aptitude Tests in Arts and Music,"
California Journal of Educational Research, VIII (1957),
A3-A5; and J. Goldstein, "Will Your Students Succeed in
Music," Etude, LXVIII (1950), 16—17.

 

31Vide E. A. Fleishman, "Predicting Code Proficiency
of Radio—Telegraphers by Means of Aural Tests," Journal of
Applied Psychology, XXXIX (1955), 150-155; and E. A.
Fleishman, M. M. Roberts, and M. P. Friedman, "A Factor
Analysis of Aptitude and Proficiency Measures in Radio-
tifiegfiaphy," Journal of Applied Psychology, LXVII (1963),
3 -3 9.

 

 

 

32T. Damgaard, "Auditory Activity and Discrimi-
nation Differences as Factors in Spelling Competence"
(unpublished Ph.D. dissertation, Florida State Uni—
versity, 1958).

33H. S. Edmondson, "The Seashore Measures of Musi-
cal Talents as a Prognostic Guide in Language Rehabili-
tation for Persons with Aphasia" (unpublished Ph.D.
dissertation, University of Michigan, 195A).

22

Insofar as has been possible to determine, it has never
been applied to the problem of excellence in acting.
This study would hope to be able to point to the rele-

vance of musical talent, as measured by the Seashore

 

Measures, to that problem.
The history of intelligence testing is a long one,

dating back seventy years to Ebbinghaus' Completion Test

 

in 1897. The first really practical scale for testing
intelligence level and expressing it in numerical units
was developed by the French psychologist Alfred Binet in
1905.3LI

The problem of measuring human intelligence proved
a compelling one, and has remained a compelling one to

35 It has given rise to one of the

the present day.
largest literatures within the general field of psychology.
It is a literature, moreover, which ranges in scope and

quality from David Engler's How to Raise Your Child's IQ36

 

to Wallace Kennedy, Vernon Van De Riet, and James White's

 

3“P. E. Vernon, Intelligence and Attainment Tests
(New York: Philos0phical Library, 1961), p. 9. This
work has an excellent introduction to the history of
intelligence testing.

35e.g., the recent inquiries of American Mensa--
a society of the tOp 2 per cent of high scorers on in-
telligence tests-—into the basic nature of intelligence,
as reported in Kansas City FOR. U. Ms. (Kansas City, Mo.:
Privately Printed, 1966).

 

 

36David Engler, How to Raise Your Child's IQ
(New York: Criterion Books, 1958).

 

23

The Standardization of the 1960 Revision of the Stanford—

 

Binet Intelligence Scale on Negro Elementary—school

 

Children in the Southeastern United States.37

 

Binet's work, which was directed primarily toward
the detection of feeble—mindedness, was extensively re-
vised in 1916 by L. M. Terman of Stanford University to
cover a much wider range of intelligence than Binet's
original battery. Terman, working with the suggestions
of the German psychologist Wilhelm Stern, also intro-

duced the idea of the Intelligence Quotient. In brief,

 

the Stanford—Binet scale measures mental age. This mental

age is then divided by the subject's chronological age,

and the quotient is multiplied by 100 to give the sub-

ject's I. Q.38
It is important to note that the idea of I. Q.,

which has had both supporters and detractors among psycholo-

gists,39 is of little importance to this study. While it

is possible to calculate a subject's I. Q. from the re-

sults of the Intelligence testing instrument used in

 

37W. Kennedy, V. Van De Riet, and James White, Th3
Standardization of the 1960 Revision of the Stanford-Binet
Intelligence Scale on Negro Elementary-school Children in
the Southeastern United States (Tallahassee: Florida
State University Press, 1961).

38

 

Vernon, 0p. cit., pp. 10—11.

39Vide Paul L. Boynton, Intelligence, Its Mani-
festations and Measurement (New York: D. Appleton and
00-, 19337; pp- 49-50.

 

 

2A

this study, the Otis Quick Scoring Mental Ability Test,

 

Gamma Fm, such a calculation has not been made. Rather,

the researcher has used the Otis Test as a measure of an

 

individual's ability to think in a particular manner or
pattern. He has, therefore, disregarded the notion of

I. Q. and simply used the Otis Test raw scores as a

 

numerical basis for comparison between and among the
groups tested for the particular thinking ability which
the test measures. The comparison attempts to determine
the relation of this ability to ability in acting.

The procedure described above is not without
precedent. The Binet test was developed as an individual
testing instrument in which ". . . the child gives his
answers in his own words and the tester assesses their

"LAO

adequacy. The Otis Test, on the other hand, was

 

developed as a group test, and came out of the work of

Arthur Otis on the Army Alpha and Army Beta tests, which

 

 

were the first large-scale attempts at group intelligence

testing.“l Vernon notes, ".

the scoring of group
tests is objective--a matter of totaling the correct re-
sponses," and, "Army Alpha scores were not usually trans-

lated into Mentages or I. Q.'s, but graded arbitrarily on

 

a letter scale from A to E."142
A0
Vernon, op. cit., p. 17.
A1 A2

Ibid. Ibid.

 

25

The Otis Test was chosen for this study from the

 

many tests available in the intelligence testing area for
several reasons. First, a group test rather than an in-
dividual test was required because of the size of the

sample. The Otis Test is a group test which is highly

 

thought of in the field, as evidenced by the following
comment from Boynton:

The Otis Self-Administering tests are among the
most satisfactory of the present group tests for
at least four reasons. First, in the ordinary
group where the scholastic background for all the
individuals is approximately the same [a condition
of the sample of this study] the validity of these
tests as measured by scholarship prediction is
approximately as high as, or perhaps equally as
high as any of the other intelligence examinations.
Second, the tests have a very high reliability.“3

Boynton's third and fourth reasons are that the Otis Test

 

is easy to administer and easy to score. The second rea-

son for using the Otis Test in this study was that a test

 

which took the minimum amount of time to complete was
required in order to keep the imposition on the subjects'

time low. The Otis Test takes less time than any com-

 

parable test to complete. Finally, it is inexpensive
both in money cost and in time for scoring.

The general area of mechanical ability and aptitude
testing is somewhat more narrow than that of intelligence
testing, possibly because it has not appealed so strongly

to the public fancy. It tends to contain material of a

 

u3Boynton, 0p. cit., pp. 2A2—2A3.

26

more strictly technical nature. There are, for example,
no books on "How to Improve Your Child's Mechanical
Abilities Quotient." The literature does, however, have
considerable breadth, and runs from Glen Grimsley's
"Draftsmen Aptitude Tests Cut Turnover"uu to C. E.
Thompson's "Motor and Mechanical Abilities in Profes-

A5

sional Schools." For an introduction to the litera-

ture in this area, the interested reader is directed to
the work of McDaniel and Reynolds in Educational and
A6

Psychological Measurement.

 

Since the actor's creation is limited by his ability
to control his physical instrument, mechanical ability
would seem to relate to excellence in acting. Mechanical
aptitude is essentially a physiological phenomenon. Like

the Seashore Measures, the MacQuarrie Test has been tested

 

 

and validated both by its author,“7 and by other.

 

M4Glen Grimsley, "Draftsmen Aptitude Tests Cut
Turnover," Western Industry, (January 19AA).

 

“SC. E. Thompson, "Motor and Mechanical Abilities
in Professional Schools," Journal of Applied Psychology,
xxv (19u6), 2u—37.

u6J. W. McDaniel and William A. Reynolds, "A

Study of the Use of Mechanical Aptitude Tests in the
Selection of Trainees for Mechanical Occupations,"
Educational and Psychological Measurement, IV (19AA),
191-197.

 

 

”7T. W. MacQuarrie, "A Mechanical Ability Test,"
Journal of Personnel Research, V (1927), 329—337.

27

A8

researchers. Its value has been determined as a

A9

prognostic instrument in both education and industry.

Moreover, the instrument has been used for studies as

diverse as predicting aptitude for mechanics and typists,50

examining the relationship of mechanical ability to in-

51 52

telligence, guiding the mentally defective and

 

u8Vide H. Babcock and M. R. Emerson, "An Analytical
Study of the MacQuarrie Test for Mechanical Ability,"
Journal of Educational Research, XXX (1938), 50-55; R. L.
Chapman, "The MacQuarrie Test for Mechanical Ability,"
Psychometrika, XIII (19A8), 175-179; H. S. Curtis, "A
Statistical Study of the MacQuarrie Test for Mechanical
Ability" (unpublished Ph.D. dissertation, University of
Michigan, 19A2); and W. Harrell, "A Factor Analysis of
Mechanical Ability Tests," Psychological Bulletin, XXXVI

(1939), 52A.

ugVide O. Garretson, "Relationships Between Ex-
pressed Preferences and Curricular Abilities of Ninth
Grade Boys," Journal of Educational Research, XXIII
(1931), 12A-132; and E. Ghiselli, "Tests for the
Selection of Inspector-Packers," Psychological Bulletin,
XXXVIII (19A6), 586-595.

 

 

 

50W. Harrell and R. Faubion, "Selection of Tests
for Aviation Mechanics," Journal of Consulting Psychology,
IV (19AO), 10A-105; and J. M. Overhaltzer, "A Study of
the Possibilities of Predicting Typing Ability" (unpub-
lisged Master's thesis, University of Southern California,
192 ).

 

51L. W. French, "Some Factors in the Relationship
of Intelligence and Mechanical Ability" (unpublished
Master's thesis, Colombia University, 1936); and L. W.
Murphy, "The Relation Between Mechanical Ability Tests
and Verbal and Non-verbal Intelligence Tests," Journal
of Psychology, 11 (1936), 353—366. """"—

 

52E. T. Burr, "Vocational Guidance of Mental De-
fectives," Psychological Clinic, XX (1932), 55—6A.

 

28

determining scientific aptitude.53 A complete biblio-

graphy of studies relating to the MacQuarrie Test is

 

available to the interested reader from its publishers,

the California Test Bureau.5u

The MacQuarrie Test has never been applied to the

 

problem of excellence in acting. This study would hope
to show the degree to which mechanical ability as mea-

sured by the MacQuarrie Test relates to acting ability.

 

The literature in the area of personality and per-
sonality testing is as large as, or larger than, the
literature in intelligence testing. The interested reader

who wishes a general background in the field might be re-
1,55

56

ferred to either Anastas or to G. W. Allport's Pattern

and Growth in Personality. Even confining oneself to

research which makes use of the personality measurement

 

53M. Mercer, "An Analysis of the Factors of Scien-
tific Aptitude as Indicated by Success in Engineering
Curricula" (unpublished Master's thesis, Pennsylvania
State College, 1938).

5“J. C. Durna, MacQuarrie Test for Mechanical
Ability Summary of Investigations Number Two (Monterey,
California: California Test Bureau, 1950), pp. lA—l7.
This work also contains an excellent introduction to
the area of Mechanical Abilities testing.

 

 

55888 p. 19.

56G. W. Allport, Pattern and Growth in Personality
(New York: Holt, Rinehart, and Winston, 1961).

 

29

device used by this study, the California Psychological

 

Inventory, one is faced with studies as diverse as Up:

 

married Mothers, by 0. Vincent,57 and "The Stability Over

 

Time and Under Stress of Conscious and Unconscious Mascu—
linity-Feminity," by L. M. Lansky.58 For a complete
listing of studies using and concerning the CPI, the
interested reader is directed to the bibliography of

the Manual for the California Psychological Inventory.59

 

The C§I_Was developed by Dr. Harrison G. Gough as
a convenient, large-scale testing instrument to measure
personality elements which ". . . have a wide and per-
vasive applicability to human behavior, and which in
addition are related to the favorable and positive aspects
of personality rather than to the morbid and pathologi-

60

cal." Each of the eighteen individual scales which

make up the inventory has been tested for individual

 

57C. Vincent, Unmarried Mothers (New York: Free
Press of Glencoe, 1961).

 

58L. M. Lansky, "The Stability Over Time and Under
Stress of Conscious and Unconscious Masculinity-Femininity,"
American Psychologist, XVII (1962), 302-303.

59H. G. Gough, Manual for the California Psycho—
logical Inventory(Palo Alto, California: Consulting
Psychologists Press, Inc., 196A), pp. 26-33.

60Ibid., p. 5.

 

 

 

 

3O

61 Each individual scale is

validity and reliability.
regarded as a separate factor for this study.

The inventory has been applied to studies as
disparate as the prediction of professional success,
the prediction of scholastic success,63 the study of

national differences,6" and the study of parent-child

65 It has not yet been applied to the

relationships.
problem of assessing the personality of persons who achieve
excellence in acting. This study attempts to make such

application.

 

61lbid., pp. 19-2A. Also see J. V. Mitchell and
J. Pierce-Jones, "A Factor Analysis of Gough's California
Psychological Inventory," Journal of Consulting Psychology,
XXIV (1960), u53—u56.

62T. Pouncey, "Psychological Correlates of Journal-
ism Training Completion" (unpublished Ph.D. dissertation,
University of Minnesota, 195A); and L. J. Carleton, "A
Study of the Relationship of Rated Effectiveness of School
Administrators and Certain of Their Personality and Per—
sonal Background Characteristics" (unpublished Ph.D.
dissertation, University of Oregon, 1956).

63H. G. Gough, "A First Report on the Use of the
California Psychological Inventory to Predict Grades in
High School," American Psychologist, VIII (1953), 501;
and M. B. Fink, "Self—concept as It Relates to Academic
Under-Achievement" (unpublished Ph.D. dissertation,
University of California, Berkeley, 1961).

6"J. C. Brenglemann, "Differences in Questionnaire
Responses Between English and German Nationals," Acta
Psychology, XVI (1959), 339-355.

65D. E. Payne and P. Mussen, "Parent-child Relations
and Father Identification Among Adolescent Boys," Journal
of Abnormal Psychology, LII (1956), 358-362.

31

There is, of course, no literature surrounding the
Stock Index of Solipsism, since the test was developed
specifically for use in this study. The Ipdgy was
developed both because there was no existing instrument
which measured solipsism, or the tendency to view the
world as happening through self, and to broaden the
author's ability in test development. It is essentially
a test of a personality factor which utilizes a sub-
jective response to obtain a numerical rating of the
testee's tendency toward that personality factor. The
response device used is sentence completion. The test
thus utilizes the same testing theory as other, more
general, measures of personality like the Thematic

Aperception Test or the Rorschach. For a full discussion

 

 

of this theory, the interested reader is directed to

Anstey.66 The development of the SIS is discussed in

Chapter III.

 

66See p. 20.

CHAPTER III

SUBJECTS, INSTRUMENTATION, AND

PROCEDURES

Subjects

The data on which this study is based were drawn
from a sample of 100 persons. Of these 100, eighty-
seven were students and faculty members of Michigan State
University. The remaining thirteen were members of the
professional acting company of the Trinity Square Reper-
tory Theatre of Providence, Rhode Island. Care was taken
to insure that the thirteen persons from the Trinity com-
pany had the experience of a university education to
avoid bias in the sample.

The 100 persons in the sample were distributed into
four separate observational groups; actors of proven
worth and ability, beginning actors with large potential,
beginning actors with small potential, and non-actors.
Males and females were equal, or nearly equal, in number
in each group. The four groups were tested under uniform
conditions, providing data which were compared and con-

trasted statistically both within and among groups.

32

33

Ten of the actors of proven worth and ability were
drawn from the Performing Arts Company of Michigan State
University. Two were drawn from the Michigan State
faculty in acting. The final thirteen were drawn from
the Trinity Square professional company.

The ten Performing Arts Company actors were un-
animously judged to be of significantly high quality by
the theatre faculty of Michigan State University.1 The
thirteen professional actors from the Trinity company
were screened for membership in that company from among
professional applicants by the directoral staff of the
Trinity theatre in a series of rigorous auditions.

Several operations were undertaken to guard against
bias being introduced by the fact that the proven actor
group was taken from two geographically remote populations.
The proven actors from Michigan State and those from the
Trinity Square Company were treated as two sub-samples,
one of twelve persons, and one of thirteen. Means and
rank sums were calculated for the two sub-samples, and
tested for significant difference for each factor. Inter-
val data were tested by the 2 test, and ordinal data were

tested by the Mann Whitney U Test.2 Criterion level was

 

held at alpha .05.

 

1The technical and design faculty was not asked to
participate in the judging since their expertise does not
normally include skill in the judging of actors.

2Statistical sources and procedures are discussed
on pp.69-7l.

3A

There was no difference significant at alpha .05
or greater on any of the interval data. The critical
value of t was 2.069 or greater. The following were the

values for p for the individual factors tested:

TABLE l.—-Va1ues for t for the individual factors tested.

 

 

Factor Value of t
Pitch 1.95363
Loudness 0.11811
Rhythm O.A6500
Time 1.06528
Timbre 0.66913
Tonal Memory 1.A6016

General Intelligence 0.6132A
Tracing 1.A9631
Tapping 0.29392
Dotting 1.16050
Copying 0.73738
Location 1.12573
Blocks 0.27099
Pursuit O.A3723
MacQuarrie Sum 1.06257
RAT Creativity O.A6198

 

Difference significant at the alpps .05 or greater
level existed on six of the factors yielding non-parametric
data. These factors were all part of data tested by the
California Psychological Inventory. They are the following:
Dominance, Well-being, Responsibility, Communality, Achieve-
ment via Conformance, and Flexibility. While this differ-
ence is further discussed in Chapter IV, it should be

noted here that in each case where significant difference

35

existed, the difference between the actors working in
the university and those working in the professional
theatre was in the same direction as the difference be-
tween the university actor sample and the non-actor
sample. In the simplest of terms, the actors from the
professional theatre tended to be what the university
actors were, only more so. This trend becomes quite
comprehensible when one considers that the factors in
question were all psychological—social in nature, and
that the actors working in the profession and living the

vie de boheme are not subject to the same restrictions

 

as are those working in the university. Thus, actors
tended to be less dominant than non-actors, and the pro—
fessional actors tended to be less dominant than the uni-
versity actors; actors tended to be lower in communality
than non-actors, and the professional actors tended to be
even lower than the university actors, etc.

Critical value for the Mann-Whitney U at EAEE§.-05
was A1 or less. The values for U for the individual
factors tested are shown in Table 2.

The twenty—five actors with large potential were
drawn from a population of persons judged to be of large
potential by the faculty in acting at Michigan State.

All twenty-five were judged both on the basis of class

work and performance.

 

36

TABLE 2.—-Va1ues for U for the individual factors tested.

 

 

Factor Value of U
Dominance 33.500*
Capacity for Status 66.000
Sociability 7A.000
Social Presence 77.500
Self Acceptance 52.500
Well—being A0.000*
Responsibility l7.000*
Socialization A8.000
Self-control 57.000
Tolerance 51.000
Good Impression 68.500
Communality 29.500*
Achievement by

Conformance A0.500*
Achievement by

Independence 68.500
Intellectual

Efficiency 70.000
Psychological

Mindedness 65.000
Flexibility A0.000*
Femininity 52.000
Solipsism 77.000

 

*Significant values.

The twenty-five actors with small potential were
drawn from a population of persons judged by the faculty
in acting at Michigan State to be deficient in acting
talent potential. Eighteen were judged on the basis of
classroom work in acting, while the remaining seven were
judged on the basis of performance in plays at Michigan
State.

It is recognized that the assigning of these per-

sons to one sample group or the other is made largely on

37

the basis of subjective decisions. It must be pointed
out that at present the only basis available for differ-
entiation between high and low potential in acting is a
subjective decision by an experienced producer or
director. This is equally true of the differentiation
between two experienced actors for a particular role.

Since these subjective decisions regarding relative
excellence form the basis of theatrical selection practices,
there seemed to be little choice but to allow them to form
the basis of the categorization of subjects for this study.
Care has been taken to insist that the judges make their
choices on the basis of the person in question's being
clearly on the parameters of a condition of excellence;

being without qpestion of either high or low potential.

 

Persons whose potential was in question were not selected
for the sample. It might be remarked that perhaps our
very systems of decision—making in the field of acting
are part of what this study is testing.3
The only possible definition of a non-actor seemed
to be a person who had never had the experience of acting

in a play, and who had expressed no desire to do so. The

control group for this study was, therefore, drawn from

 

3For example, in choosing the preliminary obser-
vational group for the actors of proven ability, this
researcher compiled a list of thirty persons who, in
his opinion, had proven their ability in acting. The
judges who determined the composition of this obser—
vational group were in agreement at the alpha .05 level
of significance on only twelve of the original thirty.

38

the general student population of Michigan State Uni-
versity, with care taken to assure that the persons
selected had never acted, and had no wish to do so.
These persons provide the norm against which the re-
sults of the other three groups are measured, although
consideration was given to the normative information
available for four of the six testing instruments which

the study uses.

Instrumentation

 

The six tests which were used to gather data for
this study will be discussed in the order in which they
were administered to the sample. Five of the six were
chosen on the basis of their being among the best instru-
ments in the area in which the existence of factors was
being tested. The sixth test was developed by the author
specifically for this study in the absence of a reliable
testing instrument in the factor area of solipsism.
Broadly speaking, the six areas under study are the
following:

1. The ability to hear musical sounds and patterns.

2. General intelligence as measured by standard

"intelligence" testing instruments.

3. Mechanical ability and dexterity.

Creative thinking ability.

U'I

Personality make—up.

O‘\

Solipsism or "I-mindedness."

39

The following test instruments were used to measure
differences in these areas between the observational
groups:

1. The Seashore Measures of Musical Talents

2. The Otis Quick-Scoring Mental Ability Testsi

Gamma Fm

 

3. The MacQuarrie Test for Mechanical Ability
A The Remote Associates Test

5. The California Psychological Inventopy

6

. The Stock Index of Solipsism

 

Each subdivision of each test is regarded as a single
factor.

The literature surrounding these six instruments,
including that with reference to their past use, was dis-
cussed in the preceding chapter. Their development,
reliability, and validity will be discussed here.

The Seashore Measures of Musical Talents was developed
in 1917 as a practical measure of an individual's ability to
accurately differentiate between six different types of
aural stimuli relating to music. Pitch, loudness, rhythm,
time, timbre, and tonal memory are the areas in which
stimuli are presented. As a predictive instrument in
music, the test is based on the hypothesis that the
ability to differentiate accurately in these areas is

essential to competent musicianship. This study examines

 

A0

the relation of ability to differentiate in these areas
to observed excellence and non-excellence in acting.

The original set of stimuli were revised in 1939
in the presence of more accurate laboratory instruments
for the production of sound stimuli. This revision is
the one currently available. The sets of stimuli are
recorded on a 33 1/3 rpm. high fidelity phonograph re-
cord, and are presented by playing that record on a
phonograph capable of reproducing the stimuli without
distortion. The six tests are on separate bands of the
record, and are administered separately. The stimuli
were originally recorded under laboratory conditions on
magnetic tape. This tape was then edited to eliminate
imperfections, and the edited tape recorded on a master
disc from which c0pies are struck for distribution and
testing. Subjects indicate their responses on specially
designed IBM answer sheets.

In the test for pitch, the subject is asked to
determine whether the second tone is higher or lower than
the first in fifty pairs of tones. The stimuli were re-
corded from a beat-frequency oscillator through a circuit
which insures pure tones free from harmonics and over-
tones. The tones vary around a 500 cycle frequency with
an individual duration of six-tenths of a second. The
following table indicates the frequency differences be-

tween the pairs.

Al

TABLE 3.--Frequency differences between the pairs--test

 

 

for pitch.
Item Differences in
Number Cycles

1- 5 17

6-12 12
13-22 8

23-32 5

33-A0 A

Al—A5 3

A6—50 2

 

In the test for loudness, the subject is asked to
determine whether the second tone of fifty pairs of tones
is stronger or weaker than the first. The stimuli were
obtained through the same apparatus as the stimuli for
the pitch test. The frequency was held constant at AAO
cycles. The following table shows the intensity differ-
ences between the pairs.

TABLE A.--Intensity differences between the pairs—-test
for loudness.

 

Item Difference in
Number Decibels

 

1— 5
6-10
11—20
21-30
31-AO
Al-50

OI—‘I—‘NMJ:
U‘IOU'IOUTO

 

A2

In the test for rhythm, the subject is asked to
determine whether thirty pairs of rhythmic patterns are
the same or different. The stimuli were recorded from
a beat-frequency oscillator set at 500 cycles at a tempo
of ninety—two quarter notes per minute. The rhythmic

patterns are shown in Table 5.

TABLE 5.--Rhythmic patterns--test for rhythm.

 

 

Item Rhythmic
Number Pattern

1—10 5 notes in 2/A time
11-20 6 notes in 3/A time
21-30 7 notes in A/A time

 

In the test for time, the subject is asked to
determine whether the second tone of fifty pairs of tones
is longer or shorter than the first. The stimuli were
obtained from the beat-frequency oscillator set at AAO
cycles. Duration was controlled by an automatic device
for which a predetermined schedule of time intervals was
prepared. The differences in duration between the pairs
is shown in Table 6.

In the test for timbre, the subject is asked to
determine whether each of fifty pairs of tones are the
same or different in timbre or tonal quality. The

stimuli were obtained with a special generator which

A3

TABLE 6.-—Differences in duration between the pairs-—
test for time.

 

 

Item Difference in
Number Seconds

1- 5 .30

6-10 .20
11—20 .15
21—30 .125
31-A0 .10
Al-AS .075
A6-50 .05

 

produces a tone with a fundamental component of 180 cycles,
and its first five overtones. For this test, the generator
varies the intensity of the third and fourth harmonics by
reciprocal alteration. The following table shows the
decibel variations from the levels in the standard tone
used in the variable tones.

TABLE 7.--Decibe1 variations from the levels in the
standard tone used in the variable tones—-test for timbre.

 

 

Item Ath 3rd
Number Harmonic Harmonic
1-10 10 9.6
11-20 8.5 A.O
21—30 7.0 2.A
31—A0 5.5 1.2

Al—SO A.O 0.7

 

AA

In the test for tonal memory, there are thirty
pairs of tonal sequences. The sequences are divided into
ten sequences of three tones each, ten of four tones each,
and ten of five tones each. In each pair, one tone is
changed in the second playing of the sequence. The sub-
ject is asked to determine which tone is changed, and to
note it by number. The stimuli were derived from a
Hammond organ, using the eighteen chromatic steps upward
from middle C, with tempo and loudness carefully con—
trolled.” I

The only requirements for the administration of the
tests are good acoustics and minimum noise disturbance in
the test room, and that the record be played on a machine
capable of reproducing the stimuli without distortion.

No subject is to sit closer than five feet from the loud
speaker. The test may be given either individually or

to groups as large as 100. The smallest group tested

for this study was two, and the largest forty-seven. The
conditions required by the test were rigorously adhered
to by the researcher.

"Reliability," in psychological testing, "is used
to indicate the trustworthiness or stability of a test

itself, apart from its representativeness or capacity

 

"The preceding six paragraphs were condensed from
developmental information in Carl E. Seashore, Manual
of Instructions and Interpretations for the Seashore
Measures of Musical Talents (New York: The Psychological
Corporation, 1960), pp. 3-A.

 

A5

for predicting anything else."5 The reliability of the
Seashore Measures was tested with internal consistency
coefficients (Kuder-Richardson formula 21). This
statistic differs from the often used Pearson {pp in
being more likely to underestimate reliability than to
overestimate it. It is, in essence, a more conservative
statistic. Reliability has been tested for grade, high
school, and college levels. Since this study is con-
cerned only with the college level subjects, only the

6

coefficients for college level students are given here.

TABLE 8.--Consistency coefficients for college level

 

 

students.

Consistency

TeSt Coefficient
Pitch .85
Loudness .7A
Rhythm .6A
Time .71
Timbre .68
Tonal Memory .83

 

 

5For a full discussion of reliability and validity,
see Philip E. Vernon, Personality Tests and Assessments
(London: Methuen & Co., 19537, pp. 21-23, from which
this citation is taken.

6Seashore, Manual of Instructions . . ., op. cit.,

A6

Validity is commonly measured against external
criteria. A test of dominance, for example, would be
tested on the basis of its ability to differentiate
persons who had experienced problems with excessive
dominance from persons who were normal in this person-

ality trait. The Seashore Measures has not been vali-

 

dated by this procedure. The senior author of the

Measures felt that such validation would be fruitless in

 

the face of the fact that there were more factors in-
volved in musical talent than those measured by the

Seashore Measures. Since this study does not attempt

 

to use the Seashore Measures as a predictive instrument,

 

but rather examines the relationship of the raw Seashore

 

data to observed excellence in acting, external validity
is of little consequence. Nevertheless, the following
statement of Opinion from Seashore himself is illumi-
hating:

The Measures have been validated for what they pur-
port to measure. . . . When we have measured the
sense of pitch, that is, pitch discrimination, in
the laboratory with high reliability, and we know
that pitch was isolated from all other factors,
no scientist will question but that we have mea-
sured pitch. . . . It is easy to show that we
cannot find a good violinist who does not have a
good sense of pitch; or a good pianist who does
not have a good sense of intensity, . . . But it
does not follow that goodness in these capacities
alone will make a good artist.7

 

7Ibid.

 

A7

Seashore felt, in other words, that the primary question
was the reliability of the tests. Since this inquiry

does not seek to relate the Seashore Measures to musical

 

talent, but rather to examine its relationship to ob-
served excellence in acting, reliability would appear
to be the prime question for this study.

In 1916, Arthur S. Otis began working on a series
of sets of analogies, opposites, and other items which
involved intellectual comprehension. His work with
these items led to the development of the first group

test of intelligence, the Otis Quick-Scoring Mental

 

Ability Tests.8 This study makes use of the 1939 re-
vision of the Gamma Fm, or adult form, of that test.

Philip Vernon, in The Measurement of Abilities, made

 

the following valuable generalization about the construc-
tion of group intelligence tests:

Group tests generally include a "battery" of some
half—dozen (between about four and twelve) "sub-
tests," each containing some twenty (between ten
and fifty) questions or items. Alternatively they
are arranged in "omnibus" form, where items of all
kinds are mixed up. In the battery type, each sub-
test has its own instructions and sample items,
and is timed separately. In the omnibus form the
instructions and samples may be printed along with
the questions, or included in a preliminary prac-
tice sheet, and one time limit suffices for the
whole test. An omnibus test is therefore easier
to give, but a test battery has the advantage of
providing rest pauses every few minutes.9

 

8Philip E. Vernon, Intelligence and Attainment Tests

(New York: Philosophical Library, 1961), p. 17.

 

9Philip E. Vernon, The Measurement of Abilities
(London: University of London Press, Ltd., 19567, p. 156.

 

 

A8

The Otis Test is in omnibus form, with instructions and

 

samples on a preliminary practice sheet. The short time
required to complete the test, thirty minutes, makes it
valuable in a study like this one which requires large
blocks of the subjects' time. Its reputation for ex-
cellence is discussed on page 25.

It has been remarked that group intelligence tests

in general, and the Otis Test in particular, tend to be

 

influenced by the scholastic level and ability of the

10

subject. Because of this fact, and because the norms

for the Otis Test extend only to the undergraduate level,

 

care was taken to insure that all subjects had completed
the freshman level of college course work. It has not
been shown that course work beyond the freshman college
level significantly affects the results of the Otis Test,
Gamma Fm.

The Otis Test, Gamma Fm is composed of eighty items
in omnibus form, and is printed on two sides of one long
piece of paper. This single sheet is specially folded
to form six letter-size pages. Pages one and two, printed
on opposite sides of the paper, contain instructions,
example questions, and an answer sheet in IBM form. Pages

three through six contain the test items.

 

10Cf. Ibid., pp. 157-158 and David Wechsler, The
Measurement of Adult Intelligence (Baltimore: Williams
and Wilkins Co., 19Al), p. 21.

A9

The subject is instructed to separate pages one
and two from the rest of the sheet. The answer sheet,
which is on page two is then placed under the remaining
sheets, and aligned with the question sheet by means of
arrows placed on the question sheet and in the appropri-
ate columns of the answer sheet. This arrangement is
meant to reduce confusion in answering. The items are
numbered from one to eighty. The subject responds by
blacking a space under the appropriate answer number on
the IBM sheet.

Only for the purpose of description, the items used
by the Otis Test, Gamma Fm are quadrachotomized into four
types: Verbal Comprehension and Reasoning, Opposites,
Spacial Reasoning, and Numerical. The following are

examples of each type from the Otis Test, Gamma Fm:

 

Verbal Comprehension and Reasoning

l. A house is most likely to have a
(1) telephone (2) doorbell (3) window
(A) carpet (5) radio.

18. Which of the following words is most like
punctuality, reliability, and honesty?
(1) intelligence (2) loyalty (3) skill
(A) divinity (5) eventuality.

Opposites

 

11. The opposite of accept is-—
(1) concept (2) take (3) forfeit
(A) reject (5) object.

58. If the following letters were placed in the
order Opposite to that in which they appear
in the alphabet, what would the fourth letter
be?

(1) Q (2) V (3) L (A) F (5) H

50

Spacial Reasoning

19. This P is to this _oas this F is to this--
(1) u. (2) 3 (3) 71 (A) _u

Numerical

 

12. A party consisted of a man and his wife,
his four sons and their wives, and three
children in each son's family. How many
were there in the party?

(1) 9 (2) l2 (3) 13 (A) 20 (5) 22

22. One number is wrong in the following series.
What should that number be?

1 A 16 6A 128 102A.
(1) 7 (2) 256 (3) 128 (A) 80 (5) 96

The Otis Test, Gamma Fm has the following numerical fre-

quency of these item types:

Verbal Comprehension Spacial
and Reasoning, OPPOSiteS R. Numerical

 

 

# Item: A2 11 ll 15

The high number of items relying on verbal comprehension
and ability may account in part for the high correlation
between scholastic ability and score on the Otis Test.

Reliability for the Otis Test, Gamma Fm has been

 

measured by coefficients of correlation between odd and
even items corrected by the Spearman-Brown formula at
.92. The mean validity index of the test items was
approximately .50, and was determined by the use of the
Flanagan table.11

The MacQuarrie Test of Mechanical Ability was first

published by Dr. Thomas W. MacQuarrie in 1925. It was

 

11Arthur S. Otis, Manual of Directions for Gamma
Test (New York: Harcourt, Brace & World, Inc., 195A7:

pp. 5-6.

51

the result of his efforts to develop a paper and pencil
test of manipulative skills which would provide stimuli
as accurate as stimuli experienced under laboratory con-
ditions, and which would also "present a situation which
would require mechanical performance and not the alert

"12 The skills

comprehension of complicated directions.
tested include hand and finger dexterity, visual acuity,
muscular control, and spacial relations, all of which

are required of the actor in using his physical instru-
ment to create a role.

The present form of the MacQuarrie Test was
develOped from an experimental version of twenty-four
parts. Each of these parts which showed a test-retest
reliability of less than .70 was rejected, reducing the
number of parts to twelve. Further research reduced the
number of parts in the test to seven. The seven parts
which make up the present form are Tapping, Tracing,
Dotting, COpying, Location, Blocks, and Pursuit. A
total score, computed as the sum of all scores divided
by three, is given in addition to the individual part
scores. The subject is given the opportunity to practice
the operation required by each test before actually per-

forming the test on which his results are recorded. A

stepwatch is used to time each individual test.

 

12Thomas W. MacQuarrie, "A Mechanical Ability
Test," Journal of Personnel Research, Vol. V, No. 9
(January 1927), 329-337.

52

In the Tracing test, the subject is asked to draw
a line through a series of staggered small openings in
a series of vertical lines. His score is the number of
lines he successfully passes without touching in fifty
seconds.

The Tapping test requires the subject to make
three dots with the point of his pencil in a series of
regularly spaced circles. He is instructed to work as
quickly as possible. His score is the number of circles
which he successfully completes in thirty seconds.

The Dotting test uses much the same testing
principle, as the Tapping test, but tests for accuracy
rather than for speed. The subject is asked to put one
dot in the middle of each of a series of very small
circles which are connected by a line. The small circles
are spaced irregularly along the connecting line. The*
dot may not touch the circumference of the circle. The
subject's score is the number of circles successfully
dotted in thirty seconds.

The Copying test requires the subject to copy a
series of simple designs into prescribed spaces. The
designs are composed entirely of straight lines. The
subject's score is the number of lines correctly drawn
in 150 seconds. Correctness is judged on the basis of

length and direction of line. The lines are judged by

53

being matched against a standard transparent overlay
provided to the tester.

In the Location test, the subject locates a series
of points in a large scale, and transposes them to an
area drawn in a smaller scale. The points are described
by letters of the Roman alphabet. The subject's score
is the number of points correctly transposed in 120
seconds.

The Blocks test presents the subject with a series
of drawings of piles of regularly sized blocks. Certain
of the blocks are marked with the letter "X". The sub-
ject is asked to determine how many of the blocks in the
pile touch each block marked with the letter "X", and to
write that number next to the letter "X". The subject's
score is the number of blocks correctly marked in 150
seconds.

The final test in the battery is the Pursuit test.
The subject is asked to follow a line through a tangle
of lines by eye alone from a beginning square to an
ending square. He identifies the ending square with a
number which was written in the beginning square. The
subject's score is the number of squares correctly
identified in 150 seconds.

The following reliabilities were determined by
Professor MacQuarrie by the test-retest method, based

on 365 cases.

 

‘pdh.ll‘.\‘.v. OUL . u"
‘~ .
4

IL

5A

TABLE 9.-—Re1iabi1ities determined by MacQuarrie by the
test-retest method, based on 365 cases.

 

 

Test Name Test-Retest Rpp
Tracing .80
Tapping .75
Dotting .7A
00pyins .86
Location .72
Blocks .80
Pursuit .76
Total Score (%) .90

 

There was a six-week interval between test and retest.
Much the same argument about validity which Sea-
shore used applies to the MacQuarrie Test.l3 While a
typist may require the same muscle control to type as she
requires to score high on the Tapping test, that does not
make the Tapping test a test of typing. If it is reliable,
the Tapping test tests the ability to make dots in circles
quickly. Thus the prime question is reliability and not
validity in this case. There have been many studies
which have examined the correlation between scores on the
subtests of the MacQuarrie battery and observed success

in work requiring mechanical ability. The most pertinent

 

13See p. A6 of this study.

 

55

of these are listed in the Summary of Investigations for

the MacQuarrie Test.“l This study seeks to make a simi-

 

 

lar comparison of score on the MacQuarrie Test and ob-
served excellence in acting.
Sarnoff Mednick first evolved and described the

theoretical background on which the Remote Associates

 

ESEE (RAT) is based in 1962.15 Briefly stated, that
theory is that the creative thinking process is one of
combining associative elements into new forms or concepts
which either meet specified requirements, or are in some
way useful. This definition does not really present any-
thing new to persons in fine arts fields. The sculptor
who takes a pair of legs from one model, a face from a
second, and a body from still a third has been creating
according to this definition. The image of a lost mother
may combine in a writer's mind with an incident read in
a newspaper to form the basis for a novel or a play.
Similarly, an actor may combine an experience from his
past life with an experience from his imagination to form
the motivational subtext for a character. The more
mutually remote the associative elements, the more
creative must be the process or the solution.

This definition suggested a basis for testing indi-

vidual differences in creative thinking. The RAT attempts

 

luSee p. 28. 15See pp. 18—19.

 

 

 

56

to use that basis in a reliable test instrument.

Mednick calls the RAT, ". . . an operational statement

of this definition," in which "the test items are in—

tended to require the examinee to perform creatively."l6
In order to form an "operational statement," it

was necessary to present stimuli from mutually remote

realities which would be familiar to all or nearly all

of the tests potential subjects. The subjects, placed

on an equal footing, could then be required to provide

the criteria—meeting missing link between them. The

need for stimuli which were common to all elements of

American culture led Mednick to use verbal materials

which make up a root part of American verbal associ-

ations. The stimuli are thus made up of word associ-

ation clusters like "ham-eggs," "kill—joy," and "red-hot."
The RAT presents thirty sets of three words. The

subject is asked to find a single word which will associate

with all three in some manner. The manner of association

differs with each stimulus—set of three words. The

following are three typical items from the RAT:

6. Sea home stomach

 

1A. note dive chair

 

16. Southern console station

 

 

l6Sarnoff A. Mednick and Martha T. Mednick,
Examiner's Manual; Remote Associates Test (Boston:
Houghton Mifflin Co., 1967), p. 1.

57

The subject is asked to write the word which relates to
the other three in the space at the right of the item.

In this case the correct responses would be "sick,"
"high," and "comfort" in that order. The subject's score
is the number of correct responses.

Reliability has been determined by the odd-even
method at .91 based on a sample of 503 persons.

There have been a number of studies which have
attempted to establish the validity of the RAT, virtually
all of which have shared the common difficulty of estab—
lishing a viable external criterion for creativity. The
two criteria most used were ratings of individuals for
their level of creativity, and ratings of creative pro-
ducts. Since individual raters may vary widely in their
judgments, some caution is advisable in attaching meaning
to the results of these studies.

Mednick and Halpern studied the relationship of
instructors' ratings of design students' creativityin
producing new designs and scores on the RAT. The corre-
lation between students' scores and faculty ratings was
.70, significant at the éiEEE.-Ol level.l7 Similarly,
Martha Mednick examined the relationship between the
ratings which graduate advisers gave their advisees on

a creativity checklist and the advisees' scores on the

 

17$. A. Mednick and S. Halpern, "Ease of Concept
Attainment as a Function of Associative Rank," Journal
of Experimental Psychology, VI (1962), 628—630.

58

RAT. She found that the two factors correlated at .55,

which was significant at sipp§_.005. This correlation

seems particularly significant in the light of the fact

that the creativity checklist had been found to have an
8

odd—even correlation of .93.1 In studies examining the

relationship of creative products to RAT scores, Dodd19

20 found that there was a considerably greater

and Gordon
correlation between creative production and RAT score

than would be indicated by chance alone. The levels of
significance which they found ranged between alphg .05

and .001.

It should be added that the BAT has not been applied
to creativity in actors. Thus the sections of the present
study which deal with the RAT are in a sense a validity
study determining how well the BAT score relates to ob-
served excellence in acting.

The California Psychological Inventory was first

issued in 1957. Its author is Harrison G. Gough. The

CPI was develOped to answer a need for a measure of a

 

l8M. Mednick, "Research Creativity in Psychology
Graduate Students," Journal of Consulting Psychology,
XXVII (1963), 265-266.

19W. E. Dodd, A Preliminary Study of Creativity
(New York: IBM Corporation, 1962).

20G. Gordon, "The Identification and Use of
Creative Abilities in Scientific Organizations,"
Seventh National Research Conference on Creativity
(Greensboro, N. C., 19667.

59

subject's general personality make-up administrable
under non-clinical conditions which was related to the
aspects of personality which relate to normal life and
social interaction.

The CPI is composed of eighteen individual scales.
A subject's score on the scales is determined by his
Different item-responses relate

responses to A80 items.

to different scales, although the test is in omnibus

form.

a special answer sheet whether an item is

as it relates to himself.

The subject is asked to respond by

 

item, "I have strong political opinions,"
to the subject's score on the Dominance scale.

false would add 0.

1. (Do) Dominance

2. (Cs) Capacity for Status
3. (Sy) Sociability

A. (Sp) Social Presence

5. (Sa) Self-Acceptance

6. (Wb) Sense of Well-being
7. (Re) Responsibility

8. (So) Socialization

9. (Sc) Self—control
10. (To) Tolerance

11. (Gi) Good Impression
12. (Cm) Communality

indicating on
true or false
Thus answering true to the
would add +1
Answering

The eighteen scales are as follows:

60

13. (Ac) Achievement by Conformance

lA. (Ai) Achievement by Independence

15. (Ie) Intellectual Efficiency

16. (Py) Psychological Mindedness

l7. (Fx) Flexibility

18. (Fe) Femininity

Each of the eighteen scales was developed indie
vidually, and was individually tested for reliability
and validity. The development, reliability testing, and
validation for each scale was then reported in a separate
article or study before the scales and items were
gathered into the present form of the CPI, It would
seem that a detailed reporting of the results of these
studies would be somewhat tedious. The following several
paragraphs, therefore, are a précis of these investi-
gations. For a complete and detailed treatment of the
research leading to the present form of the CPI, the
reader is directed to the Manual for the California
Psychological Inventory.21

The basic method of scale construction was what

"22 It consists

Dr. Gough calls the "empirical technique.
of several steps. First a criterion dimension is defined;

dominance, for example. Second, statements which seem

 

 

21Gough, Manual for the California . . ., 0p. cit.,
pp- 18-33.
22

Ibid., p. 18.

 

61

to have a bearing on the criterion dimension are made
into a preliminary battery. Third, subjects are
assembled on the basis of their being rated very high
or very low on the criterion dimension. Fourth, these
subjects take the preliminary battery. Fifth, the
battery is analyzed to determine which items are
answered in a differential manner by the subjects; that
is, which questions are consistently answered "true"
by the very high dominance persons and "false" by the
very low dominance persons. The result of these steps
is the isolation of thirty to forty items which favorably
relate to the criterion dimension.

Reliability has been determined for each of the
scales by the test-retest method. Table 10 shows the

correlations based on one study with 200 subjects.23

TABLE 10.—-Correlations based on one study with 200

 

 

subjects.

Scale Correlation Scale Correlation
Do .80 To .87
Cs .80 G1 .81
Sy .8A Cm .58
Sp .80 Ac .79
Sa .71 Ai .71
Wb .75 Ie .80
Re .73 Py .53
So .80 Fx .A9
Sc .68 Fe .73

 

 

23Ibid., p. 19.

62

The three least reliable scales, Cm, Py, and Fx, also
contain the fewest items per scale of all the scales.
They are thus more affected by a change of response to
a single item than the longer scales.

Each of the scales of the 93; has been cross-
validated against external criteria in individual
studies. In all validation studies, the correlation
between external criterion and score on the scale was
was significant at the alphs .01 level of significance
or higher.

This study seeks to use the 921 to determine if
any of the scales can be used as differentiating factors
between actors and non-actors. Each scale is treated as
an individual factor, although consideration is given to
‘the fact that the eighteen scales are divisible into four
clifferent classes of measurement. The four classes are
tzhe following:

1. Measures of poise, ascendancy, and self

assurance Do, Cs, Sy, Sp, Sa, Wb.
2. Measures of Socialization, Maturity, and
Responsibility Re, So, Sc, To, Gi, Cm.

3. Measures of Achievement Potential and
Intellectual Efficiency Ac, Ai, Ie.

A. Measures of Intellectual and Interest modes

Py, Fx, Fe.

63

It is recognized that there is something of a
case against the use of inventory instruments. The

following comment from Vernon illustrates this argu-

mentz2Ll

The majority of the questions deal with personal
matters which one might discuss with a sympathetic
and trusted friend, or a psychoanalyst, but would
certainly hesitate to commit to writing for some
relatively unknown tester to read. Many experi-
ments have in fact shown that when people do not
have to give their names they admit to larger
numbers of symptoms of maladjustment.

Because of this difficulty, and because the individual
subject's identity was not important to this study except
‘with regard to observational group membership, tests were
:marked with a code number and subjects requested not to
put their names on the answer sheets.25

The Stock Index of Solipsism (SIS) was developed
thy the author especially for use in this study. It is
figntended as a projective test of solipsism, or the ten-
Ciency of an individual to view events as happening
tlirough self since self is the only knowable thing.

The test is made up of fifty items. Each item

Cmansists of a noun accompanied by the article "the" in

tlle following form:

‘

2"Vernon, Personality Tests . . ., op. cit., p. 137.

25This code number was in the form of a date, with
tile month designating the observational group, and the

day designating the individual test. Thus, Jan. 01 -
Proven actor #1.

 

6A

22. the Kitchen

 

 

The test form is labeled "Sentence Invention Score"
to avoid any bias being introduced by the subject's know—
ing what factor is being tested.

The subject is given the following instructions
and examples:

This is a test of your ability to create complete
sentences from nouns quickly. You may use the
noun in each question below as either the subject
or the object of your sentence. Length of the
sentences is not important. In order to complete
the test as rapidly as possible, please be sure
to use the first sentence which the noun suggests
to you. Do not take time to think twice.

 

 

 

 

 

Examples

a. The wind blows the snow

b. the sun warms me

0. I see the man riding over the hill

 

 

Additional verbal instructions were given to re-enforce
the idea that the subject should use the first sentence
which the noun suggested to him, and that if he wished
he might strike out the article "the," and substitute
another word.

It was hypothesized that persons who were high in
solipsism would tend to include themselves in the sen-
tences through the words "1, me, my, mine" a signifi-
cantly greater number of times than would persons who .
were low in solipsism. A score of +1 is thus assigned
to each sentence containing the words "1, me, my," or

"mine." The subject's total score on the test is the

65

number of sentences which contain "I, me, my," or "mine."
The words "we, our" and "us," were held to include per-
sons outside the subject and were excluded from consider-
ation.

The nouns were placed on the test form in random
order, and were re—randomized to make a second form of
the test for reliability testing. The instrument was
tested for reliability by both the test-retest and odd-
even methods. The instrument was administered to forty-
five students enrolled in a specially peered class of
Speech 108, Voice and Articulation at Michigan State
University on two separate occasions and in the two
separate randomized forms. Eight days were allowed to
elapse between test and retest.

The tests were scored by hand, and the results
analyzed by the Spearman Rpp on the Michigan State Com—

puter Laboratory IBM CDC 3600 with the following result:

 

 

Correlation Significance Level
Odd-even .81 .01
Test-retest .77 .05

On the basis of these results, the decision was made that
the instrument was sufficiently reliable to be included
in the study.

The SIS has not been validated against external
criteria. Rather than attempt to establish external
criteria for solipsism, this study simply examines the

test's relation to observed excellence in acting.

I“;

 

66

Procedures

 

The six testing instruments which were used to
gather data for this study are intended for use under a
broad variety of testing conditions. The only specific
requirements were a quiet, comfortable, well-lighted
room with sufficiently good acoustics for the stimuli

of the Seashore Measures to be clearly heard in all parts

 

of the room. Uniform conditions of quiet, comfortable
writing desks, and good lighting were observed in all
rooms used for the testing. The rooms were pre-tested
to make certain that the Seashore stimuli could be heard
with equal clarity in all parts of the room.

The tests were administered to the 100 persons
comprising the sample in seven separate testing sessions.
The smallest number tested at one time was two. The
largest number tested in a single session was forty-seven.
Time required to administer all six instruments was just
slightly in excess of three hours, with some variation in
the time required to issue instructions, pick up com-
pleted tests, etc., for groups of differing sizes.

It was hypothesized that the three actor groups,
the individual members of which commonly spend blocks of
time in excess of three hours in production work requir-
ing a high degree of concentration, might complete all
six instruments at a single sitting without undue fatigue.

The control group of non-actors, who did not have this

 

 

 

67

background of production experience, were offered the
option of taking two one-and-one-half—hour sessions to
complete the six instruments to reduce the effects of
fatigue. Fourteen of the twenty-five persons in this
group took advantage of this option. The remainder
preferred to complete the six instruments in a single
session.

The six tests were administered in the following
uniform order to all subjects in the sample:

1. The Seashore Measures of Musical Talents

 

2. The Otis Quick-Scoring Mental Ability Test,
Gamma Fm

The MacQuarrie Test for Mechanical Ability

 

The Remote Associates Test

 

The California Psychological Inventory

 

mmzw

The Stock Index of Solipsism

 

The tests were administered in the above order to provide
a maximum of variety for the subject and thereby reduce
the effect of fatigue in any given test area. The re-
sponse, and the action needed to respond, to each of the
tests is quite different from the test which precedes
it and the test which follows it.

As was noted on page 63, the accuracy of personality
inventories increases if the subject does not have to put
his name on the inventory. Since the subject's indi-

vidual identity was of no importance to the study, and

 

68

since the study made use of a personality inventory, the
subjects were asked not to identify themselves in any
way on their answer sheets. Complete tests were identi-
fied as to group and subject by means of the code number.
Subjects were asked to indicate their sex by writing an
m or an f on each answer sheet. This procedure was
followed because the subject's sex makes a difference in
raw score interpretation on the California Psychological

Inventory.

 

All tests were scored by hand by the researcher. Raw
scores were used as data for all tests but the 93;. The
GE; raw scores were converted to standard scores which
may be interpreted without reference to the sex of the
subject by means of a table provided with the test in-
strument for this transformation.

The data obtained were punched onto FORTRAN state—
ment cards on a standard IBM 2A Card Punch. Two cards
were punched for each subject. The first card contained
the subject's code number,26 and the results of the first
four tests, which yield interval data. The second card

contained the code number and the results of the CPI

and SIS, which yield ordinal data. Two cards per subject

 

26This code number was derived from the code number
on the subjects' tests and their indication of their sex.
The number had four digits. The first digit (l-A) indi-
cated the observational group to which the subject be—
longed. The next two digits (01-25) indicated the subject.
The fourth digit indicated sex, with a 1 indicating male,
and a 2, female. Thus, A2Al = non-actor, #2A, a male.

 

69

were used because different statistical analyses are
necessary for ordinal and interval data.

The punched cards were then run through an IBM
A07 Accounting Machine which had been wired to print out
the contents of punched cards on paper. The contents
of the cards were verified by comparing the print-out
from the IBM A07 to the data on the completed test forms.
Mispunched cards were corrected, and the cards re-run
through the IBM A07, and the results re-checked against
the completed test forms.

The data were then submitted to a one-way analysis
of variance for each factor by two separate analysis of
variance programs on the Michigan State University Com—
puter Laboratory's IBM CDC 3600. The interval data were
analyzed by the laboratory's UNIQl program, which pro-
vides a one—way analysis of variance for data in cells
with unequal sample sizes.

The ordinal data were analyzed by a non-parametric
program prepared by the Michigan State University Com-
puter Institute for Social Science Research. This pro-
gram is largely based on the non—parametric statistics
given by Seigal.27 It is called NPAR, and is written in

28

ASA FORTRAN IV for use on the CDC 3600. The data for

 

27Sidney Seigal, Nonparametric Statistics for the
Behavioral Sciences (New York: McGraw-Hill Book Com-

pany, 1956).

28John Morris, Computer Institute for Social Sci-
ence Research Technical Report No. Al.01——Nongparametric
Statistics (East Lansing: Michigan State University,

 

70

this study were analyzed by a sub-program of the NPAR

system which provides for the use of the Kruskal-Wallis
29

 

One Way Analysis of Variance by Ranks. This sub-

 

program is designated KWAOV.3O

Normal behavioral science research design procedure
would include the presetting of a specific criterion
level. The author has chosen to discard this procedure.
It_was felt that the paucity of quantitative data on
acting, and more particularly on the difference between
actors and non—actors, obliged the researcher to report
all findings at the level of significance at which they
occurred. It must be noted, however, that conclusions
are based only on results which are significant at glpps
.10 or greater. It was felt that results at a lower
level permit the element of chance to enter too strongly
for conclusions to be drawn from them.

Interval factors which showed variance significant
at glpps .10 or greater were then further analyzed by
testing the differences between the four group means by

the 2 test.31

 

1967), p. F-l. Also see CISSR Technical Reports A0-A7.

29Seigal, op. cit., pp. 18A-193.

3O

 

Morris, op. cit., CISSR TR Al.01, p. F—2.

31Analyzed with the formula:

 

71

Ordinal factors which showed significant variance
at the slpps .10 or greater level were further analyzed
by testing for significant differences between rank sums
of the six possible permutations of the four groups.

This Operation does not give as clear an analysis Of the
cause of the variance as does a comparison of means by
the t test because the sum of the ranks differs with the
number of groups being tested. In the case of the
Kruskal-Wallis H, four groups are being tested, and
ranked, together, whereas in the case of the Mann-Whitney

32

U, two groups are ranked together. It is, therefore,
arithmetically possible to obtained two differing levels
of significance by doing the two different comparisons on
the two same groups. It is possible to obtain some in-
formation as to the cause of variance in the Kruskal-
Wallis H by the simple inspection of the four—group rank
sums. However, this information tells the source of the
variance, but not its significance. It is hoped that the
inclusion of the Mann-Whitney U_may help to clarify the
significance of the non-parametric variance for the

reader. The results of these analyses are discussed in

detail in the next chapter.

 

32Seigal, Op. Cit., pp. 18A—193.

 

YHI

 

CHAPTER IV
RESULTS

Both results and conclusions will be discussed in

 

 

‘terms of factor, rather than in terms of tests. Each Eh?
subsection of each test is regarded as a separate factor i ‘77
on those tests where subsections exist. Thus the re- 2
sults of the Seashore Measures are discussed as the g

factors of Pitch, Loudness, Time, Timbre, and Tonal
IMemory; the Otis Test results are discussed as the
factor of Intelligence; the MacQuarrie Test results are
discussed as the factors of Tapping, Tracing, Dotting,
(ZOpying, Location, Blocks, Pursuit, and Sum;l etc.
Four of the factors under consideration by this
Eitudy made a clear differentiation, significant at
Etlpha .05 or greater, between actors and non-actors.
Fkaur other factors differentiated between the four
En?oups, but only at slpps .10 or greater, and not be—
tvveen actors and non-actors. A third set of factors

diqfferentiated between actors and non—actors, but only

\_

lThe MacQuarrie Test regards the sum of the
teEstee's results, divided by three, as having a
Significance apart from the individual sub-test re—
SLllts. Please refer to pages 51-55 above.

72

73

art slpps .10 or greater. Conclusions are drawn from the
ffiirst and third of these groups in the following chapter.
Tfiie second group is dismissed as uninterpretable.

The four factors which differentiated between
Eictors and non—actors at glpps .05 or greater were
:rhythm, from the Seashore Measures, Responsibility and
(Communality from the CPI, and Solipsism from the SIS.
.Actors were higher in Rhythm, and lower in Responsibility,
Communality, and Solipsism than non—actors.

The five factors which showed difference signifi-
cant at slpps .10 or greater, but with no discernable
direction of difference were Pitch, from the Seashore
IAeasures, Tapping, Location, and Sum, from the MacQuarrie
gaggg, and Capacity for Status from the 931, In the case
(of Pitch, the high potential actors were higher than
taoth the proven actors and the non—actors. This result
:18 uninterpretable.

In the case of Tapping, Location, Sum, and CS,
true difference was between the two potential groups and
true actor and non—actor groups. There was no significant
ClL'Lfference between actors and non-actors. This would
Sllggest that the difference was due to chance, or to
SCnne hidden bias in the sample with regard to these
petrticular factors. It is obvious, since the Sum factor
155 made up from the scores on all of the other MacQuarrie
faictors, that difference on the other scores would tend

t<> influence the results of the Sum score.

 

7A»

The factors which did differentiate between
actors and non-actors, but at sipps .10 or greater, were
Tonal MemOry, from the Seashore Measures, and Self Ac-
ceptance, Socialization, and Femininity from the CPI.
Actors were higher than non-actors in Tonal Memory,
Self Acceptance, and Femininity. They were lower than
non-actors in Socialization.

The results did not indicate that measurable
differences existed in any of the other factor areas
under consideration.

Complete results for all factors may be found.in
Tables 11 through 16 on the following six:jpages. The
interpretation of these results and conclusions which
may be drawn from them are discussed in Chapter V. For
simplicity of reference, proven actors are referred to
as Group 1, high potential actors as Group 2, low

potential actors as Group 3, and non-actors as Group A.

 

L

 

TABLE ll.--Parametric analyses of variance.

75

 

 

Factor F Statistic Significance
Pitch 2.A3953 .07-
Loudness 0.61877 .60
Rhythm 5.13AA2 .002
Time 0.99235 .AO
Timbre 1.32127 .27
Tonal Memory 2.29A85 .08
Intelligence 1.12573 .3u.
Tracing 1.56081 .20
Tapping 2.AA312 .07
Dotting 1.805Al .15
COpying 1.90880 .13
Location 2.Al985 .07
Blocks 0.89lOA .AA
Pursuit 0.67500 .56
Sum 2.19538 .09
Creativity 0.21967 .88

 

 

76

TABLE l2.--Results of t-tests.

 

Group Mean

 

Factor Difference Tested 2 Significance
Pitch 2-1 = 3.6A 2.A10 .025
2—3 = 1.20 0.79A <.10*
2-A = 3.0A 2.013 .05
3-1 = 2.AA 1.615 .10
3—A = 1.8A 1.218 <.10
A-l = 0.60 0.397 <.10
Rhythm 1—2 = 2.16 2.734 .01
1-3 = 1.8A 2.329 .05
l-A = 3.00 3.797 .005
2-A = 0.8A 1.063 <.10
3—2 = 0.32 O.A05 <.10
3-A = 1.16 l.A68 .10
Tonal 1-3 = 0.8A 0.705 <.10
Memory l-A = 2.AA 2.050 .05
2-1 = O.AA 0.369 <.10
2-3 = 1.28 1.075 <.10
2-A = 2.88 2.A20 .025
3-A = 1.60 1.3AA .10
Tapping 1—A = 1.00 0.529 <.10
2-1 = 3.8A 2.031 .05
2-3 = 2.68 1.Al7 .10
2-A = A.8A 2.560 .01
3-1 = 1.16 0.613 <.10
3-A = 2.16 1.1A2 .10
Location l-A = 0.56 0.28A <.10
2-1 = 0.56 0.28A <.10
2-A = 1.12 0.568 <.10
3-1 = A.2A 2.152 .025
3-2 = 3.68 1.868 .05
3-A = A.8O 2.A36 .01
Sum 2-1 = 2.A6 0.835 <.10
2—A = 1.00 0.316 <.10
3-1 = 7.72 2.AA3 .01
3-2 = 5.08 1.607 .05
3-A = 6.08 1.92A .05
A—l = 1.6A ' 0.518 <.10

 

* <.10 indicates alpha level less than .10.

77

TABLE l3.--Non-parametric analyses of variance.

 

 

Factor H Statistic Significance
DO 3.558A .31
CS 6.7335 .08
SY 5.500A .13
SP 0.8259 .83
SA 6.8108 .08.
WE 1.0850 .78
RE 9.9655 .02_
SO 6.2852 .09
SC 0.6079 .89-
TO 2.A975 .A8
GI 0.7828 .85
CM 9.5A9A .02
AC 0.35A5 .95
AI 3.A829 .32
IE 1.9160 .59
PY 1.7970 .62
FX 5.A755 .lA
FE 7.0533 .07
SIS 11.8111 .008

M. -‘_ ‘ t!
(u

 

 

 

 

TABLE 1A.--Rank sums.

78

 

Rank Sum Group

 

 

Factor
1 2 3 A
DO 110A.00 1310.00 1A61.50 117A.50
CS *1120.50 129A.OO» 1556.50 1079.00
SY 997.50 1275.00 1A69.50~ 1308.00
SP 13A3.00 1188.00 1311.00 1208.00.
SA *1351.50 136A.50 1397.50 936.50
WS 1173.50 1202.00 1335.00 1339.50.
RE * 875.50 1A63.50 1329.00 1382.00
SO *1035.00 1261.00 1213.00 15A1.00
SC 1191.50 1308.50 1326.50 1223.50.
TO 116A.00 133A.50 1A12.00 1139.50
GI 12A0.00_ 1276.00 1355.50 1178.50
CM * 923.00. 1276.00 1302.00 15A9.00‘
AC 1268.50 1203.50 132A.50 1253.50
AI 1372.50 1351.50 1291.50 103A.50
IE. 1159.50 136A.50 1361.00 1165.00
PY 1AO7.00 1292.50 1190.00 1160.00
FX 1A00.00> 1275.00 1392.50 982.50_
FE *158A.00 1197.50 1193.50 1075.00
SIS * 933.50 1319.50 1177.50- 1619.50

 

*Indicates difference significant at alpha .10 or

greater.

 

79

TABLE 15.--Results of U tests.

 

Group

 

 

 

 

Factor Difference U Significance
cs 1-2 267.5 .19
1-3 311.5 .50,
1—4 171.0 .003
2.3 246.4 .10
2-4 233.5 .04
3—4 220.0 .03
SA 1-2 214.0 .03
1-3 198.0 .01
1-4 301.5 .41.
2—3 191.0 .009
2—4 269.5 .20
3—4 218.0 .03
RE 1-2 209.5 .02
1—3 246.0 .09-
1—4 295.0 .37
2-3 211.5 .02
2—A 188.5 .008
3-4 194.0 .01
so 1-2 156.5 .001
1—3 186.5 .007
1—4 148.5 .0007
2—3 243.5 .09
2—4 300.5 .40
3-4 272.0 .22
CM 1—2 298.5 .39
1-3 304.0 .44
1-4 310.5 .49.
2-3 274.5 .74
2-A 275.0 .14
3—A 199.5 01
FE 1—2 293.5 .36
l—3 231.0 .05
1—4 245.5 10
2-3 283.0 28 .
2—4 233.0 .06
3—4 202.0 02
SIS 1-2 200.5 .02 '
1-3 298.0 .39
1—4 234.0 .06
2-3 249.0 .10*
2—4 280.5 .26
3—4 199.0 01.

 

80

.nommomon ponpmsm

wo znppoz on use £0023 0:00» a mo monoumfixo can no zpaafinfimmoo 03p mpmowwsm mane

.msopomucoc 0cm capo: 00800000 on» :0 mmouom COOBQOO mm capos 00800000 020 CH msouom

0cm msOpom HmQOHmmomomO coozpon 080m 02p ma ooCOAOOMHU mo coauoomfio on» .nopmosw
so 0H. mamas um mumfixo oocmnommao pcOOAMHQmHm a 09033 p039 OOHOOC ommoam

 

 

0 0000 0000060 ca 0H 0000 noncoem as 0 mo. 00.00 x0
0H :03» popmopw ma 0 can» umpmosw n0 0 m0. 0m.0: 04
ma can» nopmmpw ma 0 can» poummmm ma 0 N00. 0m.mm :0
0H corp Ampmomw ma 0 swap Empmonw ma 0 m0. 00.0: m3
0H swap Ampmopw ma 0 can» powwopw ma 0 m0. 0m.mm on
ma menu umpmonw ma 0 can» gouachm ma 0 0H.v m00.a Sam
a menu Loamopw ma OH can» ampmosw ma 0 .0H.v 0mH.H coapmooq
: swap nopmogm 0H 0H can» gouachw ma 0 0H.v :mm.0 msfldome
0 swap Ampmomw ma OH p on H0300 ma 0 0H.v 000.0 espmnm
ma Oman ponmogm ma a swap Ampmopw ma 0 0H. :mm.a gonad
0 one 0H 0H 000 0H 0H 00s 0a
COHOOOLHQ coapoopfio cocOOfiMchHm m go m monomm

 

.0 00000 .npnoo m_to m 0o nnasnomuu.0a 00000

\‘HI

 

CHAPTER V

CONCLUSIONS

Some care must be taken in drawing conclusions from
the results of this study. If only because there is so
little quantitative data about the nature of acting and
the actor, there is some temptation to attach undue
significance to such data as this study uncovered.

The study attempted to establish some areas where
measurable differences might exist between actors and
the general population, and between persons Of high and
low potential in the art of acting. Because no truly
comprehensive definition of what constitutes success in
the art of acting exists, the researcher was at some
disadvantage in firmly establishing Observational cate-
gories for the study. This lack of a basic, quantifiable
term to differentiate the observational categories led
the researcher to a research design which was somewhat
"looser" than a strict behavioral scientist might desire.

The conclusions should be treated accordingly.

They should not be taken to have significance outside
the limits of the sample on which this research is

based until the results of this study are corroborated

81

 

Tfl

 

82

by further studies. While the researcher believes that
this study will satisfy the experimentalist's require—
ment of repeatability, he would prefer to regard his
results only as guideposts to further research until
the experiment has been, in fact, repeated.

There appeared to be some tendency for the pro-
fessional actors in Group 1 to score higher or lower on
some factors than their university counterparts. This
tendency was discussed in Chapter III under Subjects,
and in Chapter IV. Since the differences were almost
entirely on the section Of the CPI which deals with
social attitudes, it is possible that university life
places certain strictures on the artist. His social
behavior can not be quite as free as that of the artist
who lives and works outside the university. The person
who is able to function within the university framework
on the level on which the university subsample of Group
1 functions, of necessity, must have altered his social
attitudes sufficiently to permit him to live within the
restrictions of university life. It is probably this
softening of attitude which is reflected in the differ-

ences between the two subsamples. It is possible that

repeating this study with two Observational groups, actors

61nd non—actors, with the criterion for placement in the
Eictor group being membership in Actor's Equity, or a
CBertain mean income earned only at acting, might pro-

Vide useful results.

83

Perhaps the clearest fact which emerges from this
study is that the systems used to differentiate between
Groups 2 and 3, the students Of high and low potential,
were defective. For the most part, these two groups
showed little difference from one another, and showed
little tendency to score either in the direction Of the r
proven actors, or in the direction of the non-actors. 1
Because these two groups show no significant difference, A

they are removed from consideration in these conclusions.

 

Conclusions will be based only on the differences between V
Groups 1 and A, the proven actors and the non-actors.

It should be remarked that the very failure of the
high and low potential groups to show significant differ-
ences in factor areas where significant differences
existed between the actor and non-actor groups is not
without its own significance. The high and low potential
groups were chosen on the basis of their being on the
parameters of a condition of excellence in the opinion
of qualified judges. Put more simply, these were the
peOple who the judges were spas were either of high or
low potential. If the instruments used for this study
have any validity at all as differentiators between
actors and the general population, this lack of signifi-
cant difference between high and low potential actors
would argue that judgments about students' potential

should be made with extreme caution. On the basis of

8A

the results of this study, some doubt is raised as to
whether these judgments are accurate. One cannot help
but be reminded of the acting coach who told James Dean
that he should become an insurance salesman since he
was clearly without potential as an actor.

All of the factors under consideration in this
study did not show significant difference between actors
and non-actors. That the factors which did not show
significant difference failed to do so may be attributable
to one of, or a combination of, the following causes:

1. The factor concerned may not constitute a
measurable difference between actors and
non-actors.

2. The sample may have had some unknown bias
for the particular factor in question.

3. The testing instrument used for the factor
in question may not have been of sufficient
power and accuracy to discern existing
differences.

The eight factors from the MacQuarrie Test, and the

 

factors Intelligence, Creativity, Pitch, Loudness, Time,
Timbre, Do, Sy, Sp, Wb, Sc, To, Gi, Ac, Ai, Ie, Py, and
Fx were the factors which did not show significant

difference between actors and non-actors.

85

Five factors did show significant difference, but
not between actors and non-actors. They were Pitch,
Tapping, Location and Sum, and Cs. In the case of Pitch,
the high potential actors' scores were greater than both
the proven actors and non-actors. The proven actors
were the lowest of all. In the case Of Tapping, Lo—
cation, Sum, and Cs, the difference was between the two
potential groups on the one hand, and the actors and
non—actors on the other. There was no significant
difference between actors and non-actors. This would
suggest that the difference was due to chance, or to
some bias in the sample with regard to these particular
factors. It is Obvious that, since the Sum factor is

made up of the scores of all the other MacQuarrie factors,

 

difference on the other scores would tend to influence
the results of the Sum score. While the failure Of
most of the non—differentiating factors to show as
measurable differences is readily understandable in the
light Of the probable causes listed above, the failure
of Intelligence, Creativity, and the four Seashore
factors requires some additional comment.

The Intelligence factor was included in the study
because the researcher's Observations while working in
production led him to hypothesize that the more gifted
actor tended to be somewhat deficient in the verbal

and mathematical reasoning ability which makes up the

86

bulk of the items on most "Intelligence" tests. This
deficiency seemed to be accompanied by a corresponding
richness of emotional life which led to success in
acting. The results do not bear this hypothesis out.
That they do not do so may possibly be due to the fact
that no effort was made to discover experimentally the
degree to which actors differ with regard to intelli-
gence. It is entirely conceivable that there are at
least two distinct categories of successful actors,
one which is highly literate and has good verbal-
mathematical reasoning skills, and one which does not.
One thinks by way of example of the impression of bi-
polarity left by reading interviews with Sir Laurence
Olivier and with Paul Scofield. This is an area in
which further study is of critical importance since
verbal and mathematical reasoning play such an important
part not only in "intelligence" tests, but also in
college entrance examinations.

The failure of the Creativity factor, as measured
by the RAT, to show significant difference between actors
and non-actors is somewhat more puzzling. The evidence
Which Mednick offers as validation for his instrument
appears to be sound, and logical analysis would indicate
that his definition of creativity as the act of bring-
ing mutually remote clusters into new and useful patterns

does apply to the actor and to other artists. Indeed,

87

in taking the RAT himself in a practice session, the
researcher found that the mental sensation of finding
a "right" answer on the RAT was quite similar to the
mental sensation of finding a "right" answer to a pro-
duction problem. On this basis one might assume that
Mednick's test should have differentiated between
actors and non-actors. It did not do so.

At least a partial solution to the problem may
lie in the structure of the RAT. The mutually remote
clusters or associations which it presents are offered
as word groups in which the three known words all relate
in some common way to the unknown word. "Sea, home,"
and "Stomach" all relate to the word "sick" in the
common American word associations of "seasick, homesick,"
and "sick to your stomach." The difficulty with using
this instrument to differentiate actors from non-actors
is that the remote associations which he makes in his
creative work are non-verbal in nature. The actor's
creative work is done much more with remembered inton-
ations, images of one kind or another, and sensitivities
to psychological states than it is with words. While he
very probably does work creatively by bringing mutually
remote sets of these stimuli and reactions into new and
fruitful patterns, he does not normally work by bringing
Verbalized sets of concepts into new patterns as a

Writer, or a scientist might very well do. Research

 

~

88

should be done on the problem of finding a way to pre-
pare an instrument in which the mutually remote sets
in which a pattern must be found would relate more
directly to the stimuli with which the actor works.

It is possible that either pictorial or aural
stimuli may prove more applicable to the actor's work
than do verbal stimuli. It is also possible that the
task of bringing three unrelated words into any sort
of pattern, however fantastic, might prove more related
to acting skill than the task of finding a predetermined
"remote association" between several words. In any case,
the application of Mednick's theory of creativity to the
actor is a problem which should not be abandoned without
further study.

A cursory examination of the actor's work would
indicate that since the actor does transmit message
units about the state of his character's being through
the medium of changes in pitch, loudness, time (dur-
ation) and occasionally timbre, he should show a cor-
responding increase of skill in hearing pitches, loud-
nesses, durations, and timbres over the non-actor.

That the results of this study show that the actors
were no more skillful in differentiating one pitch,
loudness, duration, or timbre from another than were
non-actors would seem at first examination to belie

the basic premise. A careful examination of the

89

Seashore Test results offers a solution to the apparent

 

belial.

The two sections Of the Seashore Test which did

 

successfully differentiate between actors and non—actors
were Rhythm and Tonal Memory. In the first of these two

sections the testee is asked to respond to a rhythmic

J

pattern, and in the second to a pattern of pitches. He
is asked to say whether the pattern is the same or

different from the pattern which precedes it in the first

 

case, and to say where the pattern is changed in the 7

second. In the other four sections of the Seashore Test

 

the testee is asked to determine whether a specific
single tone is higher or lower than a preceding specific
single tone; whether one tone is louder or softer than
another; whether one is longer or shorter than another,
and whether one timbre is different from another. In
none of these sections is he asked to differentiate one
pattern from another.

A moment's reflection about the way in which an
actor communicates a message unit about the state of
being of his character will show that it is through
patterns of pitches, loudnesses, durations, and timbres
that the message unit is carried rather than by being
able to successfully establish a particular pitch,
loudness, duration, or timbre. An actor on tour, for

example, may vary the general level of loudness of his

90

voice to suit the size of different auditoria, but he
Will usually maintain a similar pattern of changes in
loudness in all of them. Much the same is true of
pitch, duration, and possibly even timbre. It is the
pattern of change which carries the message unit, not
the specific single tone or duration.

The two Seashore factors which did show signifi-
cant difference between actors and non-actors were the
two in which the instrument required the testee to be
able to perceive a pattern of stimuli, Rhythm and Tonal
Memory. The results for these factors show that the
actors scored significantly higher on both of these
factors than did the non—actors. It seems reasonable
to conclude from this that a superior ability to differ—
entiate rhythmic patterns, and a superior ability to
remember series Of tones constitute a measurable differ-
ence between actors and non-actors. Since this is true,
further research in developing instruments to measure
the ability to perceive patterns of change in loudness,
duration, and timbre would seem imperative.

Six other factors differentiated between actors
and non-actors at a significant level. They were Re,
Cm, Sa, So, Fe, and Solipsism. All of these factors
deal with personality. The fact that they did differen-
tiate actors from non-actors would strongly suggest that
measurable differences of personality make—up exist be-

tween the actor and the non-actor.

 

91

The results for the Re factor show that actors
scored lower than non-actors, and in fact lower than
both of the potential groups. This would suggest that
actors have a deficient sense of responsibility with
respect to the norms of American society. It is im—
portant to note that the items in the QRT Re scale
refer to things about which the normal pOpulation
feels a sense of responsibility, and do not extend to
the actor's craft. It might be suggested that the
responsibility which the actor feels to his art impairs
his "normal" sense Of responsibility. The student actor
Who cuts classes in order to rest for a university
theatre production is not an unfamiliar phenomenon.

The results for the Communality factor show that
actors were significantly lower in this factor than non-
actors. The Cm scale Of the QRT is made up of items
which normally correlate their scores highly with the
mode established for the general pOpulation on the en-
tire QRT, While it must be recognized that the 0m scale
has the third lowest reliability rating, .58, of any of
the QRT scales, this result would add further weight to
the argument that the actor's personality is measurably
different from the non—actor's.

The results for the Sa factor show that all three
Of the actor groups were higher on this factor than were
the non-actors. This would lead one to conclude that

actors do have a greater acceptance of themselves than

92

do non—actors. This acceptance Of self would probably
lead them, if the suggestions in the QRT manual are
correct about this scale, to be more self—assured and
at ease than non-actors. This result and this con—
clusion would also seem to be in accordance with the
fact that one normally assumes that a person who enjoys
performing before other peOple is a reasonably self-
assured person.

The results for the So factor show that the actors
were remarkably lower in this factor than the non-actors.
This scale is intended as a measure of social maturity.
Persons who score low on the SO scale tend to be "given
to excess, exhibition, and ostentation in their behavior,"
according to the CPI manual.l If one can ignore the
somewhat negative tone Of the manual's comment, the
need to display himself, to draw attention to himself to
a degree beyond the norm of American society, is very
probably one of the actor's driving needs. It would
appear to be a necessary abnormality. Without this
drive, the actor would hardly be likely to expose his
inner being on the stage in the way that makes exciting
theatre.

The results for the Fe factor show that the

actors were higher in this factor than were the non-actors.

 

lGough, Manual for the California . . ., Op. cit.,
p. 10.

 

93

The CPI manual gives the name Femininity to this scale
for the sake of convenience. It is actually intended to

assess the femininity or masculinity of interests rather

 

than femininity or masculinity in any sexual sense.
Items on this scale tend to center around a commitment
to activities and attitudes which are considered to be
in the province of the female by most American males.
Thus an expressed interest in mechanical things is taken
to be a masculine interest, and an expressed interest in
decoration is taken to be a feminine interest. It is
Obvious that by his very commitment to an artistic pro-
fession, the actor will be more interested in things
which the instrument regards as being feminine. It must
be noted here that the two groups were equal, or nearly
equal, in the number Of males and the number of females
in them. Thus the Fe result was not skewed by a high
number of females.

Further research is needed on all Of the ART factors
to isolate the items which contribute to the individual
scales which showed significant difference between actors
and non-actors, and to test these items in isolation from
the rest of the instrument to determine if they can be
used to differentiate actors from non-actors without the
influence of the remainder of the instrument working on
them. It might very well be of value to test other

personality inventories such as the Minnesota Multiphasic

 

 

9A

Inventory to determine whether they have scales which

 

differentiate between actors and non-actors.

Results for the Solipsism factor showed that
actors were far lower than non-actors. Since the instru-
ment tests the tendency to view the world as happening
through oneself by counting the number of times which
the testee includes himself in sentences in a non- E
directed situation, it would seem reasonable to conclude

that actors see the world as happening less through them-

 

selves than do non-actors. This conclusion would seem to r
fit logically with the fact that an actor must of neces-

sity adopt the point of View, the life position as it

were, of the character he is portraying. Viewing the
worhifrom.divergent points of view would probably tend

to break down the "I-it" relationship which the solipsist

maintains with the world at large. Corroborative re-

search should be done on the RTR to see if the results

of this study can be replicated.

Research might also be carried out with profit in
determining if the basic form of the RTR, a projective
sentence completion test, can be used to investigate
other facets of the actor's personality. As a specific
instance, this researcher has come to suspect that if the
sentences formed by the RTR testees were dichotomized
into "real" and "fantastic," actors would tend to write

a significantly higher number of "fantastic" sentences

95

than non-actors. A sentence like, "The tree has green
leaves," would be considered "real." A sentence like
"The tree is my friend," would be considered "fantastic."
Certainly one way to corroborate the results of
this study would be to extract the items, scales, and

tests which did differentiate actors from non-actors from

 

%
the body of the six tests used in this study and repeat

the study using them alone. If such a study were under—

taken and did prove that the results of this study were

repeatable, a good next step might be to administer this '

battery of items, scales, and tests to a series Of enter-
ing classes of acting majors, and correlate their scores
with individual ratings of their performances as seniors.
The factor area which was specifically excluded
from this study, medico—physiological factors, might be
examined with profit. Having, for example, blood samples
from a group of actors and a group of non-actors analyzed

by a Multiple Sequential Analyzer might reveal the exis-

 

tence Of measurable differences in blood composition.
Certainly the work which has been going on for years in
the field Of psychology in applying measures of changes
in physiology in persons Of varying occupations might be
applied to the actor.

As a final note, it has been the experience of
this researcher that certain theatre artists resist

strongly the idea that theatre art can be profitably

96

studied through quantitative methodology. These persons
seem to feel that the process of quantitative or be-
havioral analysis would inevitably take something away
from the art. This researcher would feel that the
opposite is actually true, and that only through the

use of quantitative tools can we, as scholars, arrive

at answers to the problems of theory which have for so
many years resisted solution by more subjective methods

Of analysis.

 

BIBLIOGRAPHY

97

BIBLIOGRAPHY

Books

Allport, G. W. Pattern and Growth in Personality. New
York: Holt, Rinehart, and Winston, 1961.

 

Anastasi, A. Psychological Testing. New York:
Macmillan, 1961.

 

Anstey, E. Psychological Tests. London: Nelson, 1966.

 

Blunt, J. The Composite Art of Acting. New York:
Macmillan, 1966.

 

Boynton, P. L. Intelligence, Its Manifestations and
Measurement. New York: D. Appleton and Co.,
1933.

Ching, J. Performer and Audience. Oxford: Hall Pub-
lishers, 19A7.

 

 

 

Cole, T. (ed.) Acting, A Handbook of the Stanislavsky
Method. New York: Lear Publishers, 19A7.

 

Cole, T., and H. Chinoy (eds.) Actors on Acting. New
York: Crown Publishers, 19A9.

 

Dodd, W. E. A Preliminary Study Of Creativity. New York:
IBM Corporation, 1962.

 

Durna, J. C. MacQuarrie Test for Mechanical Ability,
Summary of Investigations Number Two. Monterey,
Calif.: California Test Bureau, 1950.

 

 

Engler, D. How to Raise Your Child's IQ. New York:
Criterion Books, 1958.

 

Gough, H. G. Manual for the California Psychological
Inventory. Palo Alto, Calif." Consulting
Psychologists Press, 196A.

 

 

98

99

Kennedy, W., V. Van De Riet, and J. White. The Standardi—
zation of the 1960 Revision of the Stanford-Binet
Intelligence Scale on Negro Elementary-school
Children in the Southeastern United States.
Taélahassee: Florida State University Press,

19 1.

Lane, Y. The Psychology Of the Actor. New York: John
Day Company, 1960.

Maritain, J. Creative Intuition in Art and Poetry. New
York: Pantheon Books, 1953.

Mason, J. How to Be a More Creative Executive. New
York: McGraw-Hill, 1960.

Mednick, S. A., and M. T. Mednick. Examiner's Manual;
Remote Associates Test. Boston: Houghton Mifflin
Co., 1967.

Morris, J. Computer Institute for Social Science Re-
search Technical Report No.7Al.01--Non-parametric
Statistics. East Lansing, Mich.: Michigan State
University, 1967.

 

Oates, W., and E. O'Neill. The Complete Greek Drama.
2 Vol. New York: Random House, 1938.

Otis, A. S. Manual of Directions for Gamma Test. New
York: Harcourt, Brace, and World, Inc., 195A.

. The Performing ArtsJ Problems and Prospects.
New York: McGraw-Hill, 1965.

 

Seashore, C. E. Manual of Instructions and Interpre-
tations for the Seashore Measures of Musical Talents.
New York: The Psychological Corporation, 1960.

Seigal, S. Nonparametric Statistics for the Behavioral
Sciences. New York: McGraw—Hill, 1956.

 

Stein, M. I. Survey of the Psychological Literature in
the Area of Creativity. New York: Macmillan,
1962.

Vernon, P. E. Intelligence and Attainment Tests. New
York: Philosophical Library, 1961.

. The Measurement Of Abilities. London:
University of London Press, Ltd., 1956.

 

100

Vernon, P. E. Personality Tests and Assessments. London:
Methuen & Company, 1953.

 

Vincent, 0. Unmarried Mothers. New York: The Free
Press of Glencoe, 1961.

 

Wechsler, D. The Measurement of Adult Intelligence.
Baltimore: Williams and Wilkins Co., 19Al.

 

Articles

 

Babcock, H., and M. R. Emerson. "An Analytical Study
Of the MacQuarrie Test for Mechanical Ability,"
Journal of Educational Research, XXX (January
1938), 50-55.

 

Brenglemann, J. C. "Differences in Questionnaire
Responses Between English and German Nationals,"
Acta Psychology. XVI (1959), 339-355.

Burr, E. T. "Vocational Guidance of Mental Defectives,"
Psychological Clinic, XX (1931), 55-6A.

 

Chapman, R. L. "The MacQuarrie Test for Mechanical
Ability," Psychometrika, XIII, No. 3 (September

19A8), 175-179-

 

 

Downer, A. S. "Conference on Theatre Research,"
Educational Theatre Journal, Special Issue (June

Fleishman, E. A. "Predicting Code Proficiency of Radio-
telegraphers by Means of Aural Tests," Journal of
Applied Psychology, XXXIX (1955), 150—155.

 

 

Fleishman, E. A., M. M. Roberts, and M. P. Friedman.
"A Factor Analysis Of Aptitude and Proficiency
Measures in Radio-telegraphy," Journal of Applied
Psychology, XLVII (1963), 3AA-3A9.

 

 

Freedman, J. L., and S. A. Mednick. "Ease Of Attainment
of Concepts as a Function of Response Dominance

Variance," Journal of Experimental Psychology,
LV (1958), A63-A66.

 

Garretson, 0. "Relationships Between Expressed Prefer-
ences and Curricular Abilities of Ninth Grade
Boys," Journal of Educational Research, XXIII,
NO. 2 (February 19317, 12A-l32.

 

101

Ghiselli, E. "Tests for the Selection of Inspector-
packers," Psychological Bulletin, XXXVIII
(December 19A6), 586-595.

Goldstein, J. "Will Your Students Succeed in Music?,"
Etude, LXVIII (1950), 16—17.

Gordon, G. "The Identification and Use of Creative
Abilities in Scientific Organizations," Seventh
National Conference on Creativity, Greensboro,
N. C., 1966.

Gough, H. G. "A First Report on the Use of the Cali-
fornia Psychological Inventory to Predict Grades
in High School," American Psychologist, VIII
(1953), 501.

Graham, Kenneth L. "Relationships Between Educational
and Professional Theatre; Actor Training in the
United States," Educational Theatre Journal,
Special Issue (November 1966).

Grimsley, G. "Draftsmen Aptitude Tests Cut Turnover,"
Western Industry, (January 19AA).

 

Harrell, W. "A Factor Analysis of Mechanical Ability
Tests," Psychological Bulletin, XXXVI (July, 1939),
52A.

Harrell, W., and R. Faubion. "Selection of Tests for

Aviation Mechanics," Journal Of Consulting Psychology,
IV (July, 1940), 104-105.

Kansas City FOR.4U. Ms. Kansas City, Mo.:
Privately Printed, 1906.

 

Lansky, L. M. "The Stability Over Time and Under Stress
Of Conscious and Unconscious Masculinity-Femininity,"
American Psychologist, XVII (1962), 302—303.

 

Larson, W. S. "Practical Experience with Music Tests,"
Music Education Journal, XXIV (1938), 68-7A.

 

Lewis, D., J. G. Saeveit, and C. E. Seashore. "Revision
of the Seashore Measures of Musical Talents,"
University Of Iowa Studies Aims Progress Research

No. 65. Iowa City: University of Iowa Press,
19A0.

Mabie, E. C. "The Responses of Theatre Audiences, Experi-
mental Studies," Speech Monographs, XIX, NO. A
(November 1952), 235-2A3.

102

MacQuarrie, T. "A Mechanical Ability Test," Journal of
Personnel Research, V, No. 9 (January, 1927).

 

 

McDaniel, J. W., and E. A. Reynolds. "A Study of the
Use of Mechanical Aptitude Tests in the Selection
Of Trainees for Mechanical Occupations," E92?
cational and Psychological Measurement, IV, NO.
3*Il9AA7, 191-197.

Mednick, M. "Research Creativity in Psychology Gradu—
ate Students," Journal of Consulting Psychology,
XXVII (1963), 265-266.

. "An Orientation to Research in Creativity,"
Research Memo NO. 2. Berkeley, Calif.: Uni—
versity of California, Institute of Personality
Assessment and Research, 1958.

 

Mednick, M., and S. Halpern. "Ease Of Concept Attainment
as a Function of Associative Rank," Journal of
Eyperimental Psychology, VI (1962), 6284630.

 

Mednick, M., and M. T. Mednick. "The Associative Basis
Of the Creative Process," HEW Cooperative Research
Project NO. 1073, University of Michigan, 1965.

 

Mitchell, J. V., and J. Pierce-Jones. "A Factor Analysis
of Gough's California Psychological Inventory,"
Journal Of Consulting Psychology, XXIV (1960),
A53-A56.

Murphy, L. W. "The Relation Between Mechanical Ability
Tests and Verbal and Non-verbal Intelligence Tests,"
Journal of Psychology, II (1936), 353—366.

Payne, D. E., and P. Mussen. "Parent-child Relations and
Father Identification Among Adolescent Boys,"
Journal of Abnormal Psychology, L11 (1956) 358-362.

Seagoe, M. V., and J. C. Gowan. "The Relation Between
Interest and Aptitude Tests in Arts and Music,"
California Journal of Educational Research, VIII
(1957), A3-A5.

Smith, R. G. "A Semantic Differential for Theatre Con-
cepts," Speech Monographs, XXVII, No. 1 (March
1961), 1:8.

Stock, W. H. and R. Grubbs. "MLL, A New Base for Scene
Paint," Educational Theatre Journal, XVII, NO. A
(December 1965), 237-239.

103

Thompson, C. E. "Motor and Mechanical Abilities in
Professional Schools," Journal of Applied Psychology,
XXV (November 19A6), 2A—37.

 

Whitehall, B., and F. Kodman. "A Study of Audience Re-
action to a Stereotype Character," Educational
Theatre Journal, IV, NO. 2 (December, 1952),

139-192.

 

 

Unpublished Materials

 

Alexanian, M. "Elizabeth the Queen." Unpublished Master's
thesis, Michigan State University, 1950.

Carleton, L. J. "A Study of the Relationship of the
Rated Effectiveness of School Administrators and
Certain of Their Personality and Personal Back-
ground Characteristics." Unpublished Ph.D. dis-
sertation, University of Oregon, 1956.

Curtis, H. S. "A Statistical Study of the MacQuarrie
Test for Mechanical Ability." Unpublished Ph.D.
dissertation, University of Michigan, 19A2.

Damgaard, T. "Auditory Activity and Discrimination
Differences as Factors in Spelling Competence."
Unpublished Ph.D. dissertation, Florida State Uni-
versity, 1958.

Edmondson, H. S. "The Seashore Measures of Musical
Talents as a Prognostic Guide in Language Rehabili-
tation for Persons with Aphasia." Unpublished
Ph.D. dissertation, University Of Michigan, 195A.

Farnum, S. E. "Prediction of Success in Instrumental
Music." Unpublished Ph.D. dissertation, Harvard
University, 1950.

Fink, M. B. "Self-concept as It Relates to Academic
Underachievement." Unpublished Ph.D. dissertation,
University of California, Berkeley, 1961.

French, L. "Some Factors in the Relationship of Intelli—
gence and Mechanical Ability." Unpublished Master's
thesis, Columbia University, 1936.

Kane, A. S. "A Correlative Study of Musical Aptitude and
Music Appreciation as Measured by the Seashore
Measures of Musical Talents and the Fisher Measure-
ments of Musical Appreciation." Unpublished Master's
thesis, Boston University, 1950.

10A

Karlin, J. E. "The Factorial Isolation of the Primary
Auditory Abilities." Unpublished Ph.D. disser—

tation, University of Chicago, 19A2.

Kepke, A. N. "A Study of Communication of Perception
of Character Among Actors, Director, and Audience
Using Q Methodology." Unpublished Ph.D. disser-
tation, Michigan State University, 1963.

Mercer, M. "An Analysis Of the Factors of Scientific
Aptitude as Indicated by Success in Engineering
Curricula." Unpublished Master's thesis, Pennsy-

lvania State University, 1938.

"A Study of the Possibilities of
Unpublished Master's

Overhaltzer, J. M.
1928.

Predicting Typing Ability."
thesis, University of Southern California,

Pouncey, T. "Psychological Correlates of Journalism
Training Completion." Unpublished Ph.D. disser-

tation, University Of Minnesota, 195A.

Sheffer, I. "Emotional Memory in Acting Technique."
Unpublished Master's thesis, Michigan State Uni-

versity, 1958.

APPENDIX

105

106

SENTENCE INVENTION SCORE Student Number
Form B

INSTRUCTIONS

This is a test of your ability to create complete sentences
from nouns quickly. You may use the noun in each question
below as either the subject or the Object of your sentence.
Length of the sentences is not important. In order to com—
plete the test as rapidly as possible, please be sure to
use the first sentence which the noun suggests to you.

Do not take time to think twice.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

EXAMPLES
a. The wind blows the snow
b. the sun warms me
c. I see the man riding over the hill
QUESTIONS
1. the night
2. the gas
3. the phone
A. the worm
5. the bottle
6. the pencil
7. the cane
8. the silk
9. the boat

 

 

10. the book

 

SIS
Form B

11.

 

l2.

 

l3.

 

1A.

 

15.

 

l6.

 

l7.

 

18.

 

19.

 

20.

 

21.

 

22.

 

23.

 

2A.

 

25.

 

26.

 

27.

 

107

the

the

the

the

the

the

the

the

the

the

the

the

the

the

the

the

the

whip

Page

 

shore

 

snail

 

friend

 

wine

 

carrot

 

08.1"

 

farm

 

can

 

tree

 

hole

 

kitchen

stars

 

 

mother

 

window

 

ink

 

match

 

SIS
Form B

28.

 

29.

 

30.

 

31.

 

32.

 

33.

 

3A.

 

35.

 

36.

 

37.

 

38.

 

39.

 

A0.

 

Al.

 

A2.

 

A3.

 

AA.

 

108

the

the

the

the

the

the

the

the

the

the

the

the

the

the

the

the

the

shoe

Page 3

 

cloud

 

bird

 

chair

 

woman

 

grass

 

wool

 

tub

 

roof

 

desk

 

beast

 

sea

 

bed

 

scales

 

photo

 

arm

 

children

 

SIS
Form B

A5.

 

A6.

 

”7o

 

A8.

 

A9.

 

50.

 

109

the

the

the

the

the

the

fish

Page A

 

mask

 

home

 

moon

 

glass

road

 

 

 

hum-a..— .. -

 

nICHIGAN STATE UNIV. LIBRARIES
IIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIII
31293100490279