THE EFFECTS OF NON -COMPETITIVE, INDIVIDUAL
7 COMPETITIVE, AND GROUP COMPETITIVE SIT UATIONS
ON THE VERBAL AND FIGURAL CREATIVITY OF COLLEGE
STUDENTS

Dissertation for the Degree of Ph. D.
MICHIGAN STATE UNIVERSITY
JERROLD H. ABRAMSON
.1976

 

1 .
This is to certify that the . - ‘1}. i I

thesis entitled

The effects of
non-competitive, individual competitive,
and group competitive situations on the

verbal and figural creativity

of colEgfiEégggents

Jerrold H. Abramson

has been accepted towards fulfillment
of the requirements for

Ph,D. Educational

degree in
Psychology

 

Major professor I

0-7639 I

 

 

' zuml

 

 

ABSTRACT
THE EFFECTS OF
NON-COMPETITIVE, INDIVIDUAL COMPETITIVE, AND

GROUP COMPETITIVE SITUATIONS ON
THE VERBAL AND FIGURAL CREATIVITY OF COLLEGE STUDENTS

BY

Jerrold H. Abramson

. This study was concerned with the environment most
conducive to creative productivity. Specifically, the
major objective of this investigation was to assess the
relative effects of non—competitive, individual competi-
tive, and group competitive situations on the verbal and
figural creativity of college students. An additional
objective was to examine whether sex interacted with the
competitive treatments to affect verbal and figural
creativity.

The instrument selected to measure verbal creativity
was Torrance's Product Improvement Activity. This task
provides scores on verbal fluency, verbal flexibility,
verbal originality, and verbal elaboration. The instru-
ment selected to measure figural creativity was Torrance's
Parallel Lines Activity. This task provides scores on
figural fluency, figural flexibility, figural originality,

and figural elaboration.

Jerrold H. Abramson

A sex by treatment post-test only factorial design
with an equal number of observations per cell was employed.
Paid volunteer subjects were randomly assigned to either
the non-competitive, individual competitive, or group
competitive treatment conditions. In the non-competitive
condition (n=28), subjects were told not to be concerned
about how their performance on the creativity task com-
pared to that of the other subjects working on the tasks.
In the individual competitive condition (n=28), subjects
were first divided into four mixed—sex groups. They then
were told that the individual in each group achieving the
best overall performance on the creativity tasks would
win a reward of ten dollars. In the group competitive
condition (n=28), subjects were also divided into four
mixed-sex groups. They then were told that the group
achieving the best overall performance on the creativity
tasks would win a reward of seventy dollars to be shared
equally among the members (i.e., ten dollars for each
subject).

To assess the results, two separate 2 by 3 multi-
variate analyses of variance (MANOVA's) were the major
statistical procedures utilized. One MANOVA was performed
on the scores obtained on the verbal creativity task. The
other MANOVA was performed on the scores obtained on the
figural creativity task. When significant treatment effects
were obtained, Turkey pair-wise comparison were conducted

to determine where specific treatment differences lay.

Jerrold H. Abramson

Two supplementary statistical procedures were also carried
out. First, correlation matrices both within and across
all treatment conditions were computed to determine the
intercorrelations of the measures of verbal and figural
creativity. Second, analyses of covariance were conducted
to determine the effect of the different treatments on
verbal originality and verbal elaboration when verbal
fluency was held constant.

The statistical analyses produced the following
results:

(1) Verbal creativity was differentially affected by the
treatment conditions. Specifically, subjects in the
individual competitive condition scored significantly
higher than subjects in the non-competitive condition
on verbal fluency and verbal originality and signifi-
cantly higher than subjects in both the non—competi-
tive and group competitive conditions on verbal
elaboration. On none of the measures did the non-
competitive treatment group differ from subjects in
group competition.

(2) Figural creativity was not differentially affected
by the treatment conditions. Although performance was
generally higher in the individual competitive
condition, none of the treatment differences were

significant.

Jerrold H. Abramson

(3) Sex did not interact with the treatments to affect
either verbal or figural creativity. In general, both
sexes did best in the individual competitive condition.

(4) The sexes did not differ significantly on any of the
measures of verbal creativity. However, males did
significantly better than females on two of the fig-
ural creativity measures: figural flexibility and
figural originality.

(5) Almost all of the verbal and figural creativity
measures correlated significantly with each other
(26 of the 28 intercorrelations were significant).
However, the verbal measures and the figural measures
correlated more highly among themselves than they
did with each other.

(6) Intercorrelations among the creativity measures were
generally highest in the individual competitive condi-
tion and lowest in the non-competitive condition.

(7) When verbal fluency was held constant, verbal original-
ity was no longer significantly affected by the treat-
ment conditions. However, the treatments still had a
significant effect on verbal elaboration when the
effects of verbal fluency were accounted for.

The most important implication of this study is that
individual competition, in which individuals compete against
other individuals in order to attain a reward, may be an
effective procedure for instructors to use when verbal

creativity is one of their instructional objectives.

Jerrold H. Abramson

Further research in this area is needed, however, before

more definitive and broad generalizations can be drawn.

THE EFFECTS OF
NON-COMPETITIVE, INDIVIDUAL COMPETITIVE, AND
GROUP COMPETITIVE SITUATIONS ON
THE VERBAL AND FIGURAL CREATIVITY OF COLLEGE STUDENTS

By

.I k
_ I.

Jerrold H? Abramson

A DISSERTATION

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

DOCTOR OF PHILOSOPHY

Department of Counseling, Personnel
Services, and Educational Psychology

1976

To my parents,

who have loved and believed in me always

ii

ACKNOWLEDGMENTS

My sincere appreciation is extended to Dr. Don E.
Hamachek, who served as my academic advisor and dissertation
committee chairman. He has always been unselflish with his
interest, time and support throughout my graduate career.
Knowing Don helped make my years of graduate study a most
rewarding experience.

I am grateful to Dr. James Costar, Dr. Donald M.
Johnson, and Dr. Louise Sause for serving as members of
my dissertation committee. Their support and thoughtful
comments in the formulation and completion of this disserta-
tion were much appreciated.

Special thanks are extended to two good friends,
Doug Samuels and Mark Swerdlik. Not only did they assist
me in the administration of the experiment, but they were
also always willing to lend a listening ear during many
stressful moments.

Finally, I owe my most profound gratitude to my beautiful
mom and dad. Their love gave me the strength to successfully

complete my doctoral studies.

iii

TABLE OF CONTENTS

LIST OF TABLES.

CHAPTER

I.

II.

III.

INTRODUCTION AND REVIEW OF THE LITERATURE. .

The Problem. . . . .

Importance of the Study.

Conception of Creativity . . . .

Review of Related Literature .
Procedures for stimulating creativity .
Competition and Education . . . . .
Competition and Productivity on non-

creative tasks .
Competition and Productivity on
creative tasks . . . . .
Sex differences: Reaction to individual
and group competition. . . . . . . . . .
Statement of Hypotheses. . . . . . . . . . . .

METHOD .

Experimental Design. . . .
Subjects . . . . . . . . . . . .
Instruments. . . .

Experimental Procedure

Scoring. . . . . .
Data Analysis.
Limitations. . . . .
RESULTS.

Hypothesis One

Hypothesis Two . . . . . . . .

Hypothesis Three . . . . . . .

Hypothesis Four. . .

Summary of Hypothesis Testing.

Additional Analyses. .
Effect of Sex on Verbal Creativity.
Effect of Sex on Figural Creativity .
Intercorrelations of verbal and figural

creativity measures. . . . .

iv

Page

vi

I—l

bmflmwr—A

CHAPTER

Covariate Analysis: Verbal originality
and Verbal elaboration. . .

Summary of Findings
IV. DISCUSSION. . . . . . . . . . . . . . . .

Discussion of Findings. .
Treatment effect on verbal creativity.
Treatment effect on figural creativity
Sex by Treatment interaction effects
Sex differences in verbal and figural
creativity. .
Relationships among verbal and figural
creativity measures
Suggestions for Future Research

APPENDIX A: DEMOGRAPHIC DATA SHEET.
APPENDIX B: RAW SUMMED SCORES

REFERENCES

Page

77
79
82
82
82
85
87
88

89
92

95
96

99

Table

10.

11.

12.

LIST OF TABLES

Experimental design of the study

Multivariate and Univariate Tests for Treatment
Effects on measures of Verbal Creativity .

Observed means and standard deviations of verbal
creativity measures within each treatment
condition 0 O I O O O O O O O O O O O O O

Multivariate and Univariate Tests for Sex by
Treatment Interaction Effects on measures of
Verbal Creativity. . . . . . .

Observed sex by treatment means for measures of
Verbal Creativity. . . . . . . . . . . . . .

Multivariate and Univariate Tests for Treatment
Effects on measures of Figural Creativity.

Observed means and standard deviations of
figural creativity measures within each
treatment condition. . . . . . . . . . .

Multivariate and Univariate Tests for Sex by
Treatment Interaction Effects on measures of
Figural Creativity . . . . . . . . . . .

Observed sex by treatment means for measures
of Figural Creativity. . . . . . . . . . .

Multivariate and Univariate Tests for a Sex
Effect on measures of Verbal Creativity.

Observed means of males and females for measures
of Verbal Creativity

Multivariate and Univariate Tests for a Sex
Effect on measures of Figural Creativity

vi

Page

38

58

58

61

61

63

63

66

66

71

71

72

Table Page

13. Observed means of males and females for measures
of Figural Creativity. . . . . . . . . . . . . . . 72

14. Intercorrelations of Verbal and Figural Creativity
measures across all treatment conditions . . . . . 75

15. Intercorrelations of Verbal and Figural Creativity
measures within Non-competitive Condition. . . . . 75

16. Intercorrelations of Verbal and Figural Creativity

measures within Individual Competitive Condition . 76
17. Intercorrelations of Verbal and Figural Creativity

measures within Group Competitive Condition. . . . 76
18. Analysis of Covariance: Verbal Originality. . . . 78
19. Analysis of Covariance: Verbal Elaboration. . . . 78

vii

CHAPTER I
INTRODUCTION AND REVIEW OF THE LITERATURE

The Problem

 

Creativity has been a topic of interest throughout
the history of man, although it is only within the past
two decades that it has been studied in any systematic and
scientific way. Indeed, in his presidential address to
the American Psychological Association in 1950, Guilford

noted that only 0.2% of all the titles in Psychological

 

Abstracts were in any way related to creativity. Such

 

a statistic readily indicated the appalling neglect of
this important and unique characteristic of human behavior.
With the increasing awareness of the need to identify
and develop the creative potential and abilities of our
children in order to meet the challenges of the Space Age,
creativity research has dramatically and steadily increased
since the mid-fifties. The majority of such research has
focused on identifying the nature of the creative person.
Literally hundreds of investigations have sought to deter-
mine the intellectual and personological traits that
characterize creative individuals. Despite the fact that
many of these investigations have used different procedures

and measuring instruments and have tested different types

of samples, the findings they have obtained have been
remarkably congruent with one another (Dellas G Gaier, 1970).
Research on this aspect of creativity, therefore, has been
most rewarding, for we now have a sound, empirically based
understanding of the attributes and qualities creative
individuals tend to possess.

In contrast to research on the creative person, the
empirical findings obtained in other areas of creativity
research have not been nearly as consistent nor as plenti-
ful. Particularly is this so with respect to research on
the creative environment. In reviewing many of the studies
which have examined the environmental conditions that
facilitate or inhibit creative productivity, Stein (1974)
points out that there are still many unresolved problems
in this area. One such problem which is still open to
question is whether or not competition is an effective
stimulant of creative productivity. While there is a
considerable body of evidence indicating that competition
enhances productivity on non-creative tasks (e.g., Maller,
1929; Julian & Perry, 1967; Freischlag, 1973; Clifford,
1971), very little research has examined the relationship
between competition and performance on divergent or
creative thinking tasks. An exhaustive review of the
literature reveals that only four studies to date have
specifically compared performance on creative tasks under
non-competitive and competitive conditions (Torrance, 1959;

Torrance, 1965; Raina, 1968; Adams, 1968). The results

obtained have been contradictory. While Torrance (1959,
1965) and Raina (1968) found that competition facilitated
creative production, Adams (1968) found competition to be
detrimental.

Considering the meager and contradictory evidence on
the efficacy of competition for creative achievement,
further and more detailed study in this area seems warranted.
Accordingly, the major purpose of this investigation is to
examine the effects of competition on the creative
productivity of college students, a here-to-fore unstudied
population in this area. This study will specifically
investigate the following questions:

(1) Does competition enhance creative productivity?

(2) Does individual competition have the same effect
on creative productivity as group competition?

(3) Does the effect of individual and/or group
competition on creative productivity depend upon
whether or not the creative task is verbal or figural?

(4) Does the effect of individual and/or group competition

on creative productivity depend upon the sex of the
individual?

Importance of the Study

 

The world we live in today is faced with a variety of
highly complex and difficult problems. Many, if not all,
of these problems will only be solved when creative,
original ideas are applied to them. In recognition of this
fact, many eminent scholars have urged those in the business
of education to direct more attention towards developing

and fostering the creative abilities of students. Piaget,

perhaps one of the world's most celebrated psychologists,
has asserted: "... the principal goal of education is
to create men who are capable of doing new things, not
simply repeating what other generations have done--men
who are creative, inventive, and discovers." (1964, p. 5)
Guilford (1968) has claimed that the future of mankind will
be determined by present and future efforts to understand
and control creative performances. According to him,
"Creativity is the key to education in its fullest sense
and to the solution of mankind's most serious problems"
(p. 147). Torrance (1971) is yet another psychologist
who believes that creativity is essential to human
satisfaction and fulfillment. As he sees it:

It is becoming increasingly clear that nothing

can contribute more to the mental health and

general welfare of our own nation and to the

general satisfaction of its pe0p1e than a

general raising of the level of creativity.

The stifling of creativity cuts at the very

roots of satisfaction in living and eventually
creates overwhelming tension and breakdown.

(p. 221)

If we accept the argument that there is a need to
encourage and promote creativity, it then becomes important
to specify the environmental conditions which are most
conducive to its expression. In short, the basic question
becomes: What kind of environment stimulates creativity?
By studying how individual and group competition affects
creative productivity, the present investigation will
examine one particular aspect of this basic question. It

is believed that the data obtained in this study may have

important implications for educational practice. To the
extent that individual and/or group competitive situations
are found to enhance creative productivity, it may be
advisable for instructors to arrange such situations in
their classroom when creative thinking is one of their
instructional objectives. Of course, to the extent that
competitive situations are found to inhibit creative
productivity, it may be advisable for instructors to

de-emphasize such situations.

Conception of Creativity

 

Most psychologists agree that creativity is a highly
complex, multifaceted phenomena. As such, it is not
surprising to find that conceptions of creativity vary
considerably among different investigators. In noting
this situation, Golann (1963) has commented:

Creativity has been viewed as a normally
distributed trait, an aptitude trait, an
intrapsychic process, and as a style of
life. It has been described as that which
is seen in all children, but few adults.

It has been described as that which leads

to innovation in science, performance in
fine arts, or new thoughts. Creativity has
been described as related to, or equatable
with intelligence, productivity, positive
mental health, and originality. It has been
described as being caused by self-actualiza-
tion and by sublimation and restitution of
destructive impulses. (p. 548)

Clearly there are a variety of approaches to
conceptualizing creativity. The present investigation
adOpts the approach put forth by Guilford (1959, 1967).

He conceptualizes creativity in terms of mental abilities

or traits involved in creative achievement. Through
factor analysis, Guilford has identified fluency,
flexibility, originality, and elaboration as the major
component traits related to creative thinking. Fluency
is defined as the ability to generate many ideas which
are meaningful and relevant to the task at hand. Fluent
thinkers can efficiently retrieve and produce many ideas
from their memory store in response to a problem. Flexi-
bility is the ability to shift from one mode of thought
to another, to approach a problem in diverse ways, or to
apply different principles to a problem. Flexible thinkers
are able to produce many different categories or types
of ideas. Originality is the ability to produce unique,
clever, or rare (i.e., statistically infrequent) ideas
which are, at the same time, appropriate to the task at
hand. Original thinkers can produce clever ideas which
are rarely thought of by most other individuals. Elabora-
tion is the ability to embellish or add detail to ideas
so that they are more interesting or complex. Elaborate
thinkers are able to develOp or embroider their ideas
by providing additional information over and above that
which is necessary to communicate the basic idea.

While Guilford readily concedes that these traits
are not all there is to creative thinking and that the
existence of these traits will not guarantee the production
of creative work, he does believe that their presence will

increase the chances for creative production. These

four traits--f1uency, flexibility, originality, and
elaboration--are the dimensions of creativity that are

of interest in this study. Since these four indices of
creativity may be expressed in either verbal form (i.e.,
through language) or figural form (i.e., through drawings
or pictures), there are actually eight different dependent
variables examined in this study. Specifically, verbal
fluency, verbal flexibility, verbal originality, and
verbal elaboration are the dependent variables that make
up the larger construct of verbal creativity. Similarly,
figural fluency, figural flexibility, figural originality,
and figural elaboration are the dependent variables that
make up the larger construct of figural creativity. The
instruments that will be used to measure these traits

are described in chapter II.

Review of Related Literature

In this section we will review the literature that
is pertinent to this investigation. Since the primary
interest of this study is with identifying environmental
conditions most conducive to creativity, we will first
review several studies which have investigated the
effectiveness of various procedures (other than competition)
for stimulating creative thinking. Next we will focus on
the variable of most concern to this study--competition.
We will initially touch on some of the arguments that

have been raised, for and against, the presence of

competition in our educational system. We will then
review much of the research concerned with the relation-
ship between competition and productivity on non-creative
tasks. Following this we will examine in detail the

few studies which have investigated the effects of compe-
tition on creative productivity. Finally, we will look
at some of the literature which suggests the possibility
of sex differences in responsiveness to individual and

group competition.

Procedures for stimulating creativity

 

As indicated earlier, the majority of creativity
research has concentrated on identifying the nature of
the creative person. While this research has resulted in
giving us a better understanding of such individuals,
the labeling of individuals as creative or non-creative
has had the unfortunate consequence of reinforcing the
notion that creativity is a fixed trait, i.e., that
a person has a certain fixed amount of creative ability
and that attempts to influence or alter an individual's
level of creative functioning will thereby prove to be
futile. In recent years, however, many researchers have
convincingly demonstrated that such a notion is unwarranted.
In this section we will review just a sampling of the
empirical research which shows that a variety of environ-
mental manipulations or procedures can significantly

influence the level of creative output of individuals.

One of the most consistent findings is that variations
in the instructions given to individuals can alter both
the quantity and quality of their creative performances.
Parnes and Meadow (1959) compared the effect of brain-
storming instructions with that of "evaluative" instructions
on the performance of college students on Guilford's
Unusual Uses test, a task which requires subjects to
write, within specified time limits, as many different
uses as possible for a common object (e.g., a paper clip).
Under the brainstorming instructions, subjects were told
to forget about quality and to concentrate solely on
producing as many ideas as possible. Conversely, under
the evaluative instructions, subjects were told to produce
only ideas of good quality and that bad ideas would be
penalized. The results revealed that a significantly
greater number and quality of responses were produced
under the brainstorming instruction condition than under
the evaluative condition.

Other investigators have found that another efficient
technique for enhancing creative output is to give
individuals instructions that establish a ”creative" set.
For example, Gilchrist and Taft (1972) observed that
college students increased the number and prOportion of
original responses on the Unusual Uses task when they
were simply told to "be original" than when they were
given only the standard instructions. Similarly, Manske

and Davis (1968) reported that either the number of original

10

responses or the number of practical responses to the same
task was increased depending upon whether or not subjects
were told to "try to be original" or "try to be practical".
An experiment by Levy (1968) also found that creative
responding on a word-association task was enhanced when
subjects were instructed to role play a creative person.
Apparently, role playing made the subjects feel that it
was appropriate and "permissable" to produce more unusual,
original responses.

Another procedure that has been found to be useful
for stimulating creative production is that of modeling.
While it has long been recognized that the models an
individual is exposed to can influence his behavior
(Bandura, 1969), not until very recently has the effect
of modeling on creative behavior been experimentally
tested. In one of the initial investigations on this
problem, Harris and Evans (1973) examined the effect of
different symbolic models (i.e., written protocols in
response to the Unusual Uses task) on the performance of
college students on a variety of creativity tasks. One
group of subjects were exposed to a model that presented
a variety of divergent, original responses. Another
group of subjects were exposed to a model that illustrated
convergent responses. A final group of subjects were not
exposed to any model at all. Analysis of the results
indicated that on subsequent testing on identical, similar,

and generalization creativity tasks, subjects exposed to

11

the divergent thinking model produced significantly more
divergent and original ideas than those exposed to the
convergent thinking or no model condition. In an extension
of this study, Harris and Evans (1974) sought to determine
whether exposure to a symbolic divergent thinking model
would be more beneficial than simply instructing subjects
to reSpond creatively. Again using college students as
subjects, they found that while instructions to respond
creatively enhanced performance on the Unusual Uses test over
that of a control group, subjects exposed to the divergent
thinking model produced the most creative, original ideas.
Belcher (1975) conducted a further study on the
effects of modeling on creative production. However,
in contrast to the studies of Harris and Evans, which
utilized written materials as models, Belcher examined
the effects of a videotaped model. In this study 92
fourth— and fifth-grade children were randomly assigned
to four groups. In one group the subjects viewed a model
who emitted highly original uses for tin cans. A second
group of subjects viewed the same model emitting unoriginal
uses for a tin can. A third group of subjects read a
booklet specifically designed to train children's idea
production. The final group of subjects were simply an
untreated control group. Based on the fluency and

originality scores from Torrance's Unusual Uses test, the

I2

results showed that the group that observed the model
emitting original uses performed significantly better
than the other three groups.

Still another valuable procedure for stimulating
creativity is to present rewards contingent on the
production of creative effort or work. Halpin and Halpin
(1973) studied the effects of reward on the creative
performance of 62 undergraduate students enrolled in
educational psychology courses. In this study all
subjects were initially tested on Verbal Form B of the

Torrance Tests of Creative Thinking. Seven days later

 

subjects were administered Verbal Form A of the Torrance
tests. However, on this second administration, half of

the subjects were told that they could each earn bonus
points to be added to any educational psychology test

of their choice depending upon how much improvement they
made over their scores obtained on the first testing.

The other half of the subjects simply took the tests

under standard instructions. Analysis of the data
indicated that the subjects who received reward instructions
scored significantly higher on verbal fluency, flexibility,
and originality than did the subjects in the control group.
A study by Bamber, Jose, and Boice (1975) also found that
giving monetary rewards contingent upon the production

of creative responses was effective in enhancing the

creative performance of college students.

13

Further study also indicates that delayed rewards
may be just as effective as immediate rewards for promoting
creative production. Ward, Kogan, and Pankove (1972)
gave 191 fifth-grade children two pairs of creativity
tests on two occasions (both tests were administered to
each child individually). On the second occasion, children
in one condition each received a one cent reward immediately
after each idea they produced; children in another condition
were given pennies after the task was completed (Delayed
Reward Condition); children in a third condition served
as controls. The findings revealed that while children in
the immediate and delayed reward conditions significantly
increased the number of ideas (fluency) relative to controls,
no differences in performance existed between the immediate
and delayed rewards conditions. A more comprehensive inves-
tigation by Johnson (1974) provides confirmatory evidence.
In this study 145 third-, fourth-, and fifth-graders
worked on several of Torrance's figural creative thinking
tests under an immediate reward, delayed reward, or no
reward condition. In the immediate reward condition, subjects
were told that if they "worked hard" on the tests, they
would each receive prizes immediately after they had
finished. In the delayed reward condition, subjects were
given the same instructional set except that they were
told they would have to wait one week to receive the prizes.
In the no reward condition, of course, no incentives were

offered. The results showed that subjects who received

14

immediate and delayed reward instructions scored approxi-
mately the same. However, both reward groups scored
significantly higher than the subjects who did not receive
any reward instructions on the measures of figural fluency,
flexibility, originality, and elaboration.

From the above studies it is clear that creative
production can be enhanced by a variety of environmental
manipulations or procedures. Brainstorming instructions,
instructions that establish a creative set, modeling, and
the presentation of rewards contingent upon creative effort
or production are just some of the ways that have been
found to be useful for stimulating creativity. The present
investigation is interested in determining if competition--
in which individuals and/or groups compete against each
other in order to attain a reward--is another procedure that
can beneficially influence creative production. Before we
examine the empirical research that is most directly
related to this problem, let us first consider some of the
general arguments, pro and con, that have been raised about

the presence of competition in our educational system.

Competition and Education

 

At all levels of education, from kindergarten through
graduate school, students are continually competing against
one another for various rewards or honors. Not surprisingly,
the students themselves are acutely aware of this situation.
Recent research indicates that students perceive competition

to be one of the primary characteristics of their school

15

environment (Johnson a Johnson, 1974).

While few would dispute the presence of competition
in our educational system, many educators believe that
competition is detrimental to the educational process.
Alfred Adler (1964) was very much Opposed to the emphasis
on competition in our schools because he felt it creates
an environment that fosters selflishness and disregard
for others. He wrote:

Under our present system we generally find
that when children first come to school they
are more prepared for competition than c00p-
eration; and the training in competition
continues throughout their school days.

This is a disaster for the child; and it is
hardly less of a disaster if he goes ahead
and strains to beat the other children

than if he falls behind and gives up the
struggle. In both cases he will be inter-
ested primarily in himself. (p. 163)

Cronbach (1963) and Jersild (1975) have pointed to
other negative side effects of competition. They suggest
that students who are frequently unsuccessful in competi-
tive situations often develop feelings of inadequacy and
tend to withdraw from those activities in which they have
failed to excel. Neill (1960) and Leonard (1968) have
also argued that a major disadvantage of stressing competi-
tion is that it subverts intrinsic motivation for learning
and thinking. Yet another criticism of competition is
that it creates an evaluative, judgmental atmosphere that
arouses excessive threat, induces undue anxiety, and,

thereby, impedes learning and achievement (Rogers, 1954;

Shaw, 1958). Indeed, with specific regard to creativity,

16

Rogers believes that the threat of evaluation that is
present in competitive situations breeds defensiveness,
which, in turn, is inimical to creative thinking.

Despite the many criticisms that have been leveled
against competition in the schools, there still remains
many educators who believe it to be beneficial to the
educational process. Ebel (1972), for example, notes
that exposure to competition in school is valuable because
it prepares students to c0pe with competitive situations
theywill undoubtedly face once they have finished with
school. Perhaps the best and most frequently cited reason
for utilizing competition in the schools is that it is
a highly effective means for motivating and enhancing
student achievement. Ausubel (1968) most succinctly
summarized this rationale when he prOposed: "Competition
stimulates individual effort and productivity, promotes
higher standards and aspirations, and narrows the gap
between capacity and performance." (p. 424) Before turn-
ing to the research which has directly tested the validity
of this pr0position for creative tasks, let us first
review the related research concerned with the relationship

between competition and productivity on non-creative tasks.

Competition and Productivity on non-creative tasks

 

Prior to examining the research in this area, it is
important to distinguish between two forms of competition:

individual competition and group competition. The former

17

is characterized by competition in which the competitors
are individuals, only one of whom can attain the goal I
(e.g., best performance on a task) and receive the reward.
The latter is characterized by competition in which the
competitors are groups, only one of which can attain the'
goal and share the reward. To facilitate organization of
the review, we will first examine those studies which have
compared a single competitive treatment (either individual
or group) with a single non-competitive treatment. We
will then examine those studies which have compared the
relative effectiveness of individual and group competition.

Competition vs. Non-competition. Triplett (1898)

 

was one of the first to investigate the relationship between
competition and productivity. He studied the effect of
individual competition on the rate at which 40 children,
ages 8 through 17, turned fishing reels. He found that 20
subjects attained higher rates in competition than when
performing alone, 10 subjects achieved their best speeds
when they were performing alone, and 10 subjects performed
equally well under both conditions. Triplett noted that
competition had the most beneficial effect on subjects
with slow speeds, the least effect on subjects capable of
very fast speeds, and the most harmful effect on young
children who became over-excited by the competition and
lost motor control.

Hurlock (1927) gave 155 fourth-, fifth-, and sixth-

graders a test of simple addition problems. On the basis

18

of these results, the children in each grade were divided
into two equal groups-~the control and the rivalry group.
At the beginning of the second day, the rivalry group was
divided into two equal sub-groups which competed against
each other on four subsequent arithmetic tests given on
four successive days. The control group also took four
more arithmetic tests but only with instructions to add
the examples as quickly and as accurately as possible.
Analysis of the results indicated that, in both grades
four and six, the average score of subjects in group
competition exceeded that of control subjects on every day
of the experiment except that of the initial performance.
More recent experiments have also supported the
beneficial effects of competition. Peretti (1971) examined
the performance of college students on a color-word
interference task under individual competitive and non-
competitive situations. On this task subjects were pre-v
sented with a list of 60 color-words. Each word was
the name of a color, but the color of each was different
from the name. The subjects' task was to correctly name
the colors and ignore the word-names. Peretti found that
performance on this task was significantly better when
subjects were given competitive rather than non-competitive
instructions. Also studying college students, both
Freischlag (1973) and Carment and Hodkin (1973) found
that performance on a simple perceptual-motor task was

enhanced when subjects competed against one another than

19

when they were tested in a non-competitive situation.

A number of studies indicate that competition can
even effectively stimulate the performance of mentally
retarded children. Stoneman and Keilman (1973) gave the

Maze and Dots subtests of the Factored Aptitude Series

 

to 40 educable mentally retarded children who ranged in
age from 8 to 14 years. In the Maze task the subject
must draw a line through a series of points without
touching them. In the Dots task the subject must put a
dot in each of a series of very small traingles, without
touching the sides. It was found that performance on
both of these tasks was significantly better when the
subjects were instructed to compete against each other
than when no such instructions were given. A study by
Brown, HOppler, VanDeventer, and Sontag (1973) suggests
that the reading comprehension performance of EMR
children can also be improved by competitive situations.
In this study, after baseline measurements indicated that
the Children were answering correctly only about 25% of
their reading comprehension questions, the children were
divided into teams and were informed that the teams would
compete against each other. The goal of the competition
was to see which team could correctly answer the most
comprehension questions. After several days of team
competition, subsequent testing revealed that the children
were answering an average of 75% of the comprehension

questions. They were thus doing three times better than

20

their performance prior to competition!

Clifford (1971) examined the performance of fifth
and sixth graders on a task similar to the digit-symbol
task of the Wechsler Intelligent Test under competitive
and non-competitive conditions. She found that competition
for a reward enhanced performance, but only when the
subjects believed they were similar in ability to those
whom they were competing against. It seems reasonable to
suggest that when individuals feel out-matched and thereby
feel they have little chance to receive the reward,
competition will not effectively stimulate their performance.

Individual competition vs. groupcompetition. The

 

studies reviewed in this section examine productivity

as a function of the type of competition in which an
individual participates. The first five studies conclude
that productivity is greater in individual than in group
competition; the latter three report findings supporting
the converse conclusion.

May and Doob (1937) contend that the American
educational system encourages individual achievement and
leads students to believe that their levels of aspiration
can be met by becoming "the best in the class" or "at the
t0p of the list." In reviewing the literature on competi-
tion, they conclude that "individual remuneration stimulates
a greater efficiency of work than group remuneration; the
work of an individual for himself is more efficient than

his work for a friend or for other men." (p. 38)

21

Sims (1928) observed this to be true for 81 college
students working on two different kinds of tasks: (1) sub-
stituting digits for letters and (2) reading with speed and
understanding. Individual competitors performed far better
than group competitors. In fact, the work produced
under group competition was only slightly superior to the
work produced when there was no competition associated
with the tasks.

In one of the most extensive studies in this area,
Maller (1929) had 1,538 children (grades five through
eight) in ten different schools complete tests involving
simple addition problems. They alternated between indi-
vidual competition (where the individual achieving the best
performance would get a prize) and group competition (where
the class achieving the overall best performance would get
a prize). The children also were given a free choice test
consisting of seven sheets of work which they could do
either for the group or for themselves.

The results revealed that both forms of competition
produced an increase in work output above the level of
unmotivated or practice work. However, the children
produced a great deal more work on the tests in individual
competition than in group competition. Although the
children initially worked equally hard for the class as
for themselves, as testing proceeded they increased the
amount of work produced for themselves and decreased

the amount of work produced for the class. On the free

22

choice test, where they could work either for themselves
or for the class, the children chose the former altern-
ative about three times as often as the latter.

Julian and Perry (1967) divided 157 college students
into four- and five-member teams to examine productivity
under individual competition, group competition, and pure
cooperation. These conditions were established by
announcing three different sets of grading criteria for
performance on a required laboratory exercise. The
criteria announced were as follows: (a) the highest grades
would go to individuals performing the best work, regardless
of team membership (individual competition); (b) the highest
grades would go to all members of the one group which
produced the single best performance (group competition);
and (c) the highest grades would be granted to all members
of each group achieving at least 90% of the possible per-
formance points, regardless of the performance of other
groups (pure cooperation).

The required exercise involved two discussion questions
to be answered in detail by each student. The findings
were that the highest overall quality and quantity of
performance occurred in individual competition and the
poorest overall performance occurred in pure c00peration.
Performance in both individual and group competition was
significantly higher than in pure cooperation.

Kakkar (1968) studied the performance of 384 Indian

college students on arithmetic problems under non-

23

competitive, individual competitive, and group competitive
situations. Each student worked on the problems under
each type of situation. While both competitive situations
were found to increase productivity, performance was
clearly best under the individual competitive condition.

In a study by Whittemore (1924), three groups of four
college students were given the task of using individual
types to print paragraphs from the daily press. In
successive trials, the subjects were instructed: (a) to
avoid competing with one another, (b) to attempt to beat
other group members, or (c) to work together to surpass
other groups. The results showed that all subjects
produced more work when competing than when not competing.
When the two types of competition were compared, it was
found that subjects worked slightly faster in group
competition than in individual competition while the
quality of the work did not differ.

In the course of this investigation, Whittemore
interrupted both competitive and non-competitive trails
to ask subjects what they were thinking and whether they
were conscious of competing. In analyzing the replies,
Whittemore (1925) was able to provide a partial explanation
for the increased productivity observed in the competitive
situations. During competitive trials, the subjects'
thoughts focused on some phase of their work or on the
work being accomplished by their rivals. In contrast,

when non-competitive trials were in progress, subjects had

24

a greater tendency to daydream and to think about ideas
irrelevant to the task.

In yet another study comparing individual and group
competition, Deutsch (1962) divided 50 volunteers from
an introductory psychology course into ten groups. Each
week during five weeks of experimentation, the groups
were presented with a puzzle on which they were to work
together. The group's task was to formulate a solution
for the puzzle. The only difference between the groups
was the difference in their instructions. Five groups
received instructions designed to elicit group competition:
"the grade or reward that each member received would be
the same and would be determined by the relative position
of his group in contrast with the other four similar
groups." Five groups received instructions designed to
elicit individual competition: "each member would be rated
in comparison with the efforts of the other four members
comprising his group, the grade or reward that each would
receive would be different and would be determined by the
relative contributions of each to the solution of the
problem." Analysis of the results indicated that the
productivity of group competitors on the whole was better
than that of individual competitors. Group competitors
not only solved the puzzle problems more rapidly, but also
came up with better solutions than the individual compet-

itors.

25

In an experiment by Raven and Eachus (1964), each
subject was seated at one corner Of a triangular board
and given the task of leveling the board by turning the
knob. A subject could not alter the level of his own
corner directly, but by turning the knob at his corner
could alter the corners at which the other two subjects
were seated. Twenty triads were told that the object
of the task was to see how fast the three of them could
level the board in comparison to the other triads (group
competition). Twenty other triads were told that the
object was to see which one of them could level his
own corner first (individual competition). Measures of
the time required for all three corners to be leveled
showed that triads in group competition Operated with
greater speed and efficiency than triads in individual
competition.

From the studies reviewed in this section it is
clear that individual and group competition may differen-
tially affect productivity on a particular task. While
some studies have found individual competition to be
better than group competition, other studies have obtained
Opposite findings. In accounting for these contradictory
results, Jones and Vroom (1964) point out that one must
consider the nature of the task in order to determine
whether individual or group competition will yield greater
productivity. More Specifically, Jones and Vroom note

that those studies which found individual competition to be

26

more effective utilized independent tasks, i.e., tasks on

 

which individuals worked separately. Conversely, those
studies which found group competition to be more effective

employed interdependent tasks, i.e., tasks requiring

 

individuals to work together.

Competition and Productivity on creative tasks

 

The studies reviewed in the previous section indicate
that competition can stimulate productivity on a variety
of non-creative tasks, with individual competition being
more effective for independent tasks and group competition
being more effective for interdependent tasks. Whether or
not competition is an efficacious procedure for enhancing
performance on creative tasks is a problem which few
researchers have tackled.

Torrance (1959) conducted the first known experimental
effort to demonstrate the effects of competition on
creativity. In this study first through sixth graders
worked on Torrance's Product Improvement Activity, a task
which requires subjects to suggest as many clever and
unusual ideas as they can for improving a toy stuffed
dog. While one group of children worked on the task under
non-competitive conditions, another group worked under
competitive conditions in which a prize was promised for
the individual who achieved the best performance. Analysis
of the data indicated that the competitive group produced
a significantly larger number (fluency) and greater

flexibility of responses than the non-competitive group.

27

This finding held up across all grade levels.

In response to Torrance's study, critics argued that
a brief "warm-up" or practice session would eliminate the
differences between the competitive and non-competitive
groups. Consequently, Torrance (1965) undertook another
study to test the validity of this criticism. First
through sixth graders were randomly assigned either to a
"practiced" condition, in which the class spent some time
prior to testing thinking of ways to improve a toy fire
truck, or to a competitive condition, in which the class
was not given any practice but was told that the individual
achieving the best performance on the task would receive
a reward. The task used to assess creative thinking was
identical to the one used in the first study.

The results revealed a fairly consistent tendency
for all children under the competitive condition to excel
those under the practiced condition. Specifically,
subjects in the competitive condition scored significantly
higher on fluency in the first, third, and fourth grades,
significantly higher on flexibility in the second and fourth
grades, and significantly higher on originality in the
second, third, and sixth grades. In no grade did the
children in the practiced condition exceed the children
in the competitive condition on any of the measures. Thus,
Torrance concluded that practice may reduce but does not
completely compensate for the stimulating effects of

competition.

28

A study by Raina (1968) provides cross-cultural
evidence for the efficacy of competition on creativity. In
this experiment 40 Indian students (equivalent to ninth-
graders in the American educational system) worked on
Torrance's Product Improvement Activity as well as his
Unusual Uses test, a task which requires the subject to
think of as many clever uses of a particular object as one
can. Twenty of the subjects worked on the tasks under
standardized conditions (the control group), while the
other twenty were told that the three individuals scoring
highest would each receive a monetary reward (the competi-
tive group). The results indicated that the competitive
group was significantly more fluent and flexible in their
ideas than the control group. For some unexplained reason,
Raina did not assess the originality scores of the two
groups. Consequently, it is not known whether competition
enhanced this particular component of creativity.

In contrast to the studies of Torrance and Raina,
Adams (1968) obtained findings suggesting that competition
is a detriment to creative production. In this study,
ninth grade students worked on a battery of Guilford's
divergent thinking tasks (Unusual Uses, Consequences,
Object Naming). Prior to testing, one group of subjects
were given the following non-competitive instructions in
addition to the standard instructions: "... It is not
important how well you do in comparison to others taking

the tests. Your answers will not be compared with those

29

of other students. You may feel free to take these tests
without worry as to how you do in relation to others."
Another group of subjects were given the following compet-
itive instructions in addition to the standard instructions:
"...we want to discover whether you can do better than the
other groups taking the tests, so your tests will be scored
and compared with the scores from the other groups. Try
very hard to do better than all of the other students if
you want your group to come out on top."

The results showed that the group given the non-
competitive instructions were significantly more flexible
in their ideas than the group given the competitive
instructions. Since no assessment was made of the fluency
and originality of the responses, it is not known how the
groups compared on these two measures. In any case, Adams'
findings are clearly inconsistent with those of the previous
studies. There are a number of differences, however,
between Adams' study and the studies of Torrance and Raina
which may partially account for the discrepant results.

One difference is the way in which the competition was
induced. Whereas Torrance and Raina offered a prize for
the individual achieving the best performance, Adams
simply instructed subjects to try to do better than
others--no concrete incentive was offered. A second
difference was in the type of competition that was
“utilized. Whereas Raina and Torrance employed individual

competition, Adams emphasized group competition.

30

Considering these differences, it seems plausible to
suggest that the subjects in Adams' study had less to
personally gain from competition than the subjects in the
other studies, and, as a result, they were not as motivated
to do well on the tasks. Why they should be less productive
than the non-competitive group, however, is still not
clear. Although Adams argues it is the fear of evaluation
in competitive situations that results in reduced creative
performance, the subjects in the other studies also knew
their performance would be evaluated. Thus, fear of
evaluation cannot be used to explain the divergent results.
Overall, the meager evidence available on the effects
of competition on creative productivity does not allow for
definitive conclusions. A number of limitations character-
ize the research that has thus far been undertaken in this
area. First, in two of the studies (Raina, 1968;
Adams, 1968) the originality dimension of creativity was
not even assessed and in no study was the elaboration
dimension assessed, yet these would seem to be the two
most important factors to consider. Second, no study has
examined the effects of competition on the creative
performance of college-age subjects. As Stein (1974)
suggests, the effects of competition may vary with age.
Third, no study has examined how competition affects per-
formance on non-verbal or figural creativity tasks.
Previous research indicates that while a particular

procedure may enhance verbal creativity, it may have little

31

beneficial effect on figural creativity. Torrance (1970),
for example, found that dyadic interaction effectively
stimulated performance on verbal creativity tasks, while
Brown (l973)reported that dyadic interaction hampered
productivity on figural creativity tasks. It may well be
that competition also has a differential effect on verbal
and figural creativity tasks. A final limitation of
research in this area is that no study has compared the
relative effects of individual and group competition on
creative performance. It is apparent from the studies
reviewed in the previous section, however, that these
different forms of competition may yield different results.
The present study prOposes to eliminate the limitations
that have been innumerated above. Specifically, the major
Objective of this investigation is to test whether
non-competitive, individual competitive, and group
competitive situations differentially affect the verbal
and figural creativity of college students. Verbal
fluency, verbal flexibility, verbal originality, and
verbal elaboration are the dependent variables used to
define verbal creativity. Figural fluency, figural flex-
ibility, figural originality, and figural elaboration are
the dependent variables used to define figural creativity.

Sex differences: Reaction to individual andgroup
competition

 

 

A considerable amount of research demonstrates that
competition can have a stimulating effect on the produc-

tivity of both males and females on a variety of tasks

32

(Hurlock, 1927; Peretti, 1971; Freischlag, 1973; Clifford,
1971; Torrance, 1965). Very little research, however,
has been specifically designed to test whether a particular
form of competition (i.e., individual vs. group) is more
beneficial for males than females, or vica-versa. Inspec-
tion of the literature on socialization and sex-role
standards suggests the possibility that the sexes may differ
in their responsiveness to individual and group competition.
In his comprehensive review of research on sex-typing,
Mischel (1970) notes that females in our society are social-
ized to be nurturant, dependent, and oriented to the needs of
others. Males, on the other hand, are socialized to be
independent, competitive, and self-reliant. Bardwick (1971)
similarly states that the dominant motive Of females is
affiliation with others, while the dominant motive of males
is personal achievement. To the extent that females are
"other-oriented", as the literature seems to suggest, it
might be expected that females would be better motivated
to perform by group rather than individual competition.
Conversely, to the extent that males are "self-oriented",
it might be expected that males would be better motivated
to perform by individual rather than group competition.
Research by Richman (1972) does, in fact, support these
expectations. In this study elementary school boys and girls
worked on addition problems under non-competitive, individual
competitive, and group competitive conditions. In the indi-
vidual competitive condition, a prize was promised to the

individual achieving the best performance within the group of

33

which he or she was a member. In the group competitive
condition, a prize was promised to the group which had the
best overall performance. The results indicated that per-
formance under the competitive conditions was better than
performance under the non-competitive condition for both sexes.
More interestingly, however, whereas boys did best in indi-
vidual competition, girls did best in group competition.
Since Richman's data are limited to elementary school
children working on a simple addition task, it would be
interesting to explore whether this differential reSponsive-
ness to the individual and group competitive situations could
be obtained among college students working on creative tasks.
Therefore, an additional objective of this study is to
investigate whether or not sex interacts with the competitive
treatments to affect performance on the verbal and figural

creativity tasks.

Statement of Hypotheses
To determine whether non-competitive, individual
competitive, and group competitive situations differentially
affect performance on the verbal creativity task, the follow-
ing major hypothesis and its associated sub-hypotheses, all
stated in the 3211 form, are to be tested:
Hypothesis 1: Verbal creativity is not differentially

affected by non-competitive, individual compet-
itive, and group competitive situations.

 

Sub-hypotheses:

(1a) Verbal fluency is not differentially affected by
non—competitive, individual competitive, and
group competitive situations.

 

34

(lb) Verbal flexibility is not differentially
affected by non-competitive, individual
competitive, and group competitive situations.

(lc) Verbal originality is not differentially
affected by non-competitive, individual
competitive, and group competitive situations.

(1d) Verbal elaboration is not differentially
affected by non-competitive, individual
competitive, and group competitive situations.

To determine whether sex interacts with the treatment

conditions (non-competition, individual competition, group

competition) to affect performance on the verbal creativity

task, the following major hypothesis and its associated

sub-hypotheses, all stated in the null form, are to be

tested:

Hypothesis 2: The effect of non-competitive, individual
competitive, and group competitive

situations on verbal creativity does not
vary with sex.

 

Sub-hypotheses:

 

(2a) The effect of non-competitive, individual
competitive, and group competitive situations
on verbal fluency does not vary with sex.

(2b) The effect of non-competitive, individual
competitive, and group competitive situations
on verbal flexibility does not vary with sex.

(2c) The effect of non-competitive, individual
competitive, and group competitive situations
on verbal originality does not vary with sex.

(2d) The effect of non-competitive, individual
competitive, and group competitive situations
on verbal elaboration does not vary with sex.

To determine whether non-competitive, individual

competitive, and group competitive situations differentially
affect performance on the figural creativity task, the

following major hypothesis and its associated sub-hypotheses,

35

all stated in the null form, are to be tested:

Hypothesis 3: Figural creativity is not differentially

affected by non-competitive, individual
competitive, and group competitive
situations.

Sub-hypotheses:

 

(3a)

(3b)

(3C)

(3d)

Figural fluency is not differentially affected
by non-competitive, individual competitive, and
group competitive situations.

Figural flexibility is not differentially affec-
ted by non-competitive, individual competitive,
and group competitive situations.

Figural originality is not differentially
affected by non-competitive, individual compet-
itive, and group competitive situations.

Figural elaboration is not differentially
affected by non-competitive, individual
competitive, and group competitive situations.

To determine whether sex interacts with the treatment

conditions (non-competition, individual competition, group

competition) to affect performance on the figural creativity

task, the following major hypothesis and its associated

sub-hypotheses, all stated in the null form, are to be

tested:

Hypothesis 4: The effect of non-competitive, individual

competitive, and group competitive
situations on figural creativity does not
vary with sex.

Sub-hypptheses:

 

(4a)

(4b)

The effect of non-competitive, individual
competitive, and group competitive situations
on figural fluency does not vary with sex.

The effect of non-competitive, individual
competitive, and group competitive situations
on figural flexibility does not vary with sex.

(4:)

(4d)

36

The effect of non-competitive, individual
competitive, and group competitive situations
on figural originality does not vary with sex.

The effect of non-competitive, individual
competitive, and group competitive situations
on figural elaboration does not vary with sex.

CHAPTER II

METHOD
This chapter presents the design, procedures, and
limitations of the study. The chapter is organized
according to the following sections: (1) Experimental
design; (2) Subjects; (3) Instruments; (4) Experimental
procedure; (5) Scoring; (6) Data Analysis; and (7) Limi-

tations.

Experimental Desigp

 

A 2 by 3 post-test only factorial design with an
equal number of observations per cell was utilized in
this study. The independent variables were sex (male
and female) and treatment condition (non-competition,
individual competition, group competition). Four
dimensions of verbal creativity-- verbal fluency, verbal
flexibility, verbal originality, and verbal elaboration--
and four dimensions of figural creativity-- figural
fluency, figural flexibility, figural originality, and
figural elaboration-- served as dependent variables.

The experimental design is depicted in Table 1. This
design is the same for both the verbal and figural

creativity measures.

37

38

TABLE 1. Experimental design of the study (note: FL=
fluency, FX= flexibility, O= originality, E= elab—

 

 

 

oration)
TREATMENT
, Ngnggmp, Ind Com . Gr Com . .
FL FX 0 E FL FX 0 E FL FX 8 E
SI
Sz
Male 2 n=14 n=14 n=14
S14
SEX
S1
S2
Female 2 n=14 n=14 n=14
S14

 

 

 

 

 

 

 

Subjects

Subjects in this study were students enrolled at
Michigan State University during spring term of 1976.
They were recruited by eight undergraduate students who
were hired by the investigator. The recruiters were told
that (1) they should each attempt to recruit an equal
number of males and females and (2) they would each
receive three dollars for every subject they recruited
who showed up for and participated in the study. In
addition, each recruiter was told to give the following
information about the study to prospective subjects!
(1) the date and time of the study and the location where
it was scheduled to take place; (2) that it was concerned

with creative thinking; (3) that it would only take about

39

30 minutes of their time; and (4) that they would each be
paid two dollars for participating. As a result of this
procedure for securing subjects, within two weeks the
recruiters were successful in obtaining the agreement of 98
students, 49 male and 49 female, to serve as subjects.

After a list of the names of the prospective subjects
was compiled, they were grouped according to sex and
college level (freshman, sophomore, junior, senior and
graduate). From these groups they were randomly assigned,
through use of a random numbers table, to the three treat-
ment conditions. Since each treatment was scheduled to take
place in a different room in Erickson Hall, three days
before the experiment a postcard was sent to each subject
indicating the room number he or she was to report to.

On the day of the experiment, a total of 86 subjects
actually showed up. Specifically, 29 subjects (15 male,
14 female) showed up for the non-competitive condition, 29
subjects (15 male, 14 female) showed up for the individual
competitive condition, and 28 subjects (14 male, 14 female)
showed up for the group competitive condition. Since it
was desirable to have an equal number of subjects in each
cell of the design of the study, data from two male subjects,
one from the non-competitive and the other from the indi-
vidual competitive condition, were randomly discarded and were
not considered in subsequent data analysis. A total of 84

subjects, 28 in each treatment condition, therefore

40

constituted the sample of this study.

Demographic data collected on all 84 subjects
indicates that the treatment groups were roughly equivalent
in terms of age and college class level. The average ages
of the subjects in the non-competitive, individual compet-
itive, and group competitive conditions were 20.1, 20.0,
and 19.9, respectively. With respect to college class
level, the non-competitive condition was composed of 8
freshman, 6 sophomores, 8 juniors, 5 seniors, and I
graduate student. The individual competitive condition
was composed of 8 freshman, 7 sophomores, 7 juniors, 5
seniors, and I graduate student. The group competitive
condition was composed of 8 freshman, 6 sophomores, 7
juniors, 6 seniors, and I graduate student. Overall,
then, the sample was composed of 24 freshman, 19 sopho-

mores, 22 juniors, 16 seniors, and 3 graduate students.

Instruments

 

Two tests from the Torrance Tests of Creative Think-

 

ipg (Form A) were the creativity tasks employed in this

study. The Product Improvement Activity (PIA) was used

to assess verbal creativity, while the Parallel Lines

Activity (PLA) was used to assess figural creativity.

These tasks were selected for three reasons:

(1) Both of them allow for scoring on the dimensions of
creativity that are of interest in this study-— flu-

ency, flexibility, originality, elaboration.

(2)

(3)

41

Research demonstrates high levels of inter- and intra-
scorer reliability on these tasks (usually over .90),
even when inexperienced scorers are used (Halpin 8
Halpin, 1974; Torrance, 1966). To reliably score
these tasks, all that seems to be necessary is

careful study of the scoring manual.

Although several critics (e.g., Crockenberg, 1972)
have noted that the validity of the Torrance tests

are still open to question, several studies have found
that such tests do possess some degree of validity for
use with adults. For example, Torrance (1968) found
that graduate students who scored high on these tasks
also made the most original and creative applications
of the knowledge presented in a course. Torrance and
Hansen (1965) also observed that creative business
education teachers (as identified by these tasks)
asked more provocative questions, more self-improving
questions, and more divergent ones than did their

less creative peers. In still another study, Sommers
(1961) reported that students carefully identified

by college industrial arts instructors as creative
scored significantly higher on several of Torrance's
tests (including the ones used in this study) than

did the less creative students. While these studies
suggest that the tasks employed in this study are of
some value for measuring creativity, the reader should

be cautioned that performance on these tasks is not

42

a completely valid index of creativity. The results
of the study, therefore, should be interpreted with
this caution in mind.

In summary, these tasks have been selected because
they provide measures that are of interest in this study,
they can be scored reliably, and they have demonstrated
some validity for use with adult pOpulations.' The follow-
ing is a description of both of these tasks:

Product Improvement Activity. This task requires the

 

subject to formulate suggestions for improving an object
with respect to its function as a toy. The object selected
for use is a stuffed toy elephant. This was chosen from
among the various stimuli experimented with by Torrance
because it most successfully minimizes differences in

the quality and nature of responses based on sex differences.
For group administration, a sketch of a toy elephant
accompanies the sheets with instructions and blanks for
the responses. The instructions given are:

In the middle of this page is a sketch
of a stuffed toy elephant of the kind
you can buy in most dime stores for a-
bout one or two dollars. It is about
six inches tall and weighs about a half
pound. In the spaces on this page and
the next one, list the cleverest, most
interesting, and unusual ways you can
think of for changing this toy so that
children will have more fun playing
with it. Do not worry about how much
the change would cost. Think only
about what would make it more fun to
play with as a toy.

Torrance reports that this task has been one of his most

dependable verbal tests. It is a complex test with a high

43

degree of face validity. It permits subjects to consider
ideas which they would not dare eXpress in a more serious
task.

The nature of the task allows for scoring on the traits
of fluency, flexibility, originality, and elaboration. The
fluency score is determined by counting the number of
relevant, separate responses or ideas given by the subject.
An idea is relevant if it tells how the toy elephant could
be improved as a plaything, i.e., something that would be
fun. The flexibility score is determined by the number of
different principles or approaches used in responding to
the task. An inclusive list of twenty-one general prin-
ciples is given in the scoring manual. Originality is
determined from a table of responses giving scoring weights
of 0, l, or 2. Approximately 350 reSponses are listed,
based on a tabulation of the responses of 594 subjects
from grades one through twelve. Rare responses showing
creative strength but not included in the list are scored
"2". Elaboration is determined by scoring the extent to
which the idea is spelled out or developed by counting
the details beyond those which are necessary to communicate
the basic idea. The scoring manual presents several
illustrations of elaborated responses.

Parallel Lines Activity. This task requires the subject

 

to make multiple associations to a single stimulus and to
express his ideas pictorially. Although the task normally

consists of three pages containing thirty sets Of parallel

44

lines, pilot testing indicated the advisability of adding
an additional page with 12 more sets of parallel lines.
Consequently, in this study the task consisted of four
pages containing 42 sets of parallel lines. The instruc-
tions given are:

In ten minutes see how many objects or

pictures you can make from the pairs of

straight lines below and on the next

pages. The pairs of straight lines

should be the main part of whatever you

make. With pencil or crayon add lines

to complete your picture. You can place

marks between the lines, on the lines,

and outside the 1ines--wherever you want

to make your picture. Try to think of

things that no one else will think of.

Make as many different pictures or

objects as you can and put as many ideas

as you can in each one. Make them tell

as complete and as interesting a story

as you can. Add names or titles in the

spaces provided.
The rationale of this task, as explained by Torrance, is
based upon incomplete figures (i.e., parallel lines)
creating tensions in the individual with the usual response
being to complete the figures in the simplest and easiest
way possible. Thus, the subject has to be able to handle
his tensions and delay the gratification of this impulse
in order to produce an original and elaborate set Of
figures.

Like the PIA, this task allows for scoring on the
traits of fluency, flexibility, originality, and elaboration.
Fluency is determined by counting the number of relevant
reSponses, although any repetitious responses are sub-

tracted from this total. A relevant response is defined as

45

one which contains or makes use in some way of the parallel
lines. Flexibility is determined by counting the number of
different categories into which the subject's responses can
be classified. The scoring manual lists 68 categories which
were derived from the responses of 500 subjects from kinder-
garten through college. Originality is determined from a
table of responses giving scoring weights of 0, l, 2, or 3.
Approximately 250 responses are listed, based on a tabulation
of the responses of 500 subjects from kindergarten through
college. Imaginative responses showing creative strength
but not included in the list are scored "3". Elaboration

is determined by counting the number of ideas communicated
by each drawing which are an addition to the minimum basic
idea. Examples of elaborated drawings are presented in the

manual.

Experimental Procedure

 

As previously indicated, subjects were randomly
assigned to the non-competitive, individual competitive,
and group competitive treatment conditions. Such random
assignment, of course, permits one to assume that the
three treatment groups were equivalent on all possible
variables. A pretest was therefore not necessary.

The eXperiment was conducted on April 29, 1976 in
three identical size classrooms in Erickson Hall on the
Michigan State University campus. Three male doctoral
candidates in educational psychology served as the

eXperimenters. A description of the procedure that was

46

adhered to within each treatment condition now follows:

\

Non-competitive condition- Subjects assigned to this

 

condition were seated at separate desks as they entered
the classroom. After the subjects had arrived (a total

of 29 out of a possible 32 actually showed upl), they
each filled out a demographic sheet indicating their name,
age, and college level (see Appendix A). Next, the
experimenter read the following instructions to all the
subjects:

In this study I am interested in
assessing the creative thinking
abilities of college students. You
will all be given two separate tasks

on which to work. One task is a
measure of verbal creativity, while

the other task is a measure of figural
creativity. It is important that you
be as creative as you can on both

tasks so that my results will be reli-
able. Do not worry, however, about

how others are doingyon the tasks. It
is not my purpose to compare your score
with thOse of Othersl so you don't have
to be concerned about how you do in
relatibn to others: You will allghave
ten minutes to work on each task. I
will tell you when the first ten min-
utes are up. At that time you will step
work and I will hand out the second task
you are to work on. Before you begin
work on each task, be sure to read the
instructions for each task carefully.
Are there any questions?...............
Do not ask any questions after we begin.
Just use your own best judgment if you
have a question. Also, do not talk to
anyone while you work on the tasks.

 

 

 

 

 

 

 

1

As noted earlier, data from one of these 85 was randomly
discarded following the experiment and was not considered
in data analysis.

47

After these instructions were read, the PIA was handed
out face down to each subject by the experimenter. Once
every subject had the PIA, the experimenter then told
the subjects to turn it over and to read the instructions to
the task silently while he read them aloud. This was done
in order to insure that all subjects finished reading the
instructions at the same time and would therefore have an
equal start on the task. After the subjects had worked
for ten minutes on the PIA, they were told to stop and the
PLA was then handed out by the experimenter. The same
procedure used in administering the PIA was used for the
PLA.

Upon completing work on both tasks, all the subjects
were thanked for participating and they were each paid
two dollars as they left the classroom.

Individual competitive condition- After the subjects

 

assigned to this condition had arrived (a total of 29 out

of a possible 33 actually showed upz), they were divided into
four mixed-sex groups. Two of the groups were composed

of four males and three females, another group was composed
of four females and three males, and the last group was
composed of four females and four males. Next, each

group was seated together at separate desks in a location

of the room apart from the other groups. The desks were

 

2
Again, data from one subject was randomly discarded prior
to data analysis.

48

arranged in a circle for each group. After all the subjects
were seated, they each filled out the demographic sheet
indicating their name, age, and college level. The exper—
imenter then read the following instructions to all subjects:

In this study I am interested in assessing
the creative thinking abilities of college
students. You will all be given two sep-
arate tasks on which to work. One task is
a measure of verbal creativity, while the
other task is a measure of figural creativ-
ity. Since it is important that you be as
creative as you can on both tasks so that
my results will be reliable, the individual
in each group achievinggthe best combined‘
performance on these tasks will be rewarded
with ten dollars. Therefore, try to do
better than the other members Ofyyour group
so that you can win the reward. YOu w1ll
all have ten minutes to work on each task.
I will tell you when the first ten minutes
are up. At that time you will step work
and I will hand out the second task you

are to work on. Before you begin work on
each task, be sure to read the instructions
for each task carefully. Are there any
questions?..... ........ .... DO not ask any
questions after we begin. Just use your
own best judgment if you have a question.
Also, do not talk to anyone while you work
on the tasks.

 

After these instructions were read, the PIA was

handed out face down to each subject by the experimenter.
Once every subject had the PIA, the experimenter then told
the subjects to turn it over and to read the instructions to
the task silently while he read them aloud. Following ten
minutes of work on the PIA, the subjects were told to step
and the PLA was then handed out by the experimenter. The
same procedure used in administering the PIA was used for

the PLA.

49

Upon completing work on both tasks, all the subjects
were thanked for participating and were informed that the
winner in each group would be contacted and given the ten
dollar prize within one week. Each subject was then paid
two dollars as he or she left the classroom.

Group competitive condition- After the subjects assigned

 

to this condition had arrived (a total of 28 out of a
possible 33 actually showed up), they were divided into
four mixed-sex groups. While two of the groups were
composed of four males and three females, the other two
groups were composed of four females and three males.
Next, each group was seated together at separate desks in
a location of the room apart from the other groups. The
desks were arranged in a circle for each group. After
all the subjects were seated, they each filled out the
demographic sheet indicating their name, age, and college
level. The experimenter then read the following instruc-
tions to all the subjects:

In this study I am interested in assessing
the creative thinking abilities of college
students. You will all be given two
separate tasks on which to work. One task
is a measure of verbal creativity, while

the other task is a measure of figural
creativity. Since it is important that

you be as creative as you can on both tasks
so that my results will be reliable, Egg
group achieving the best overall performance
on these tasks will be rewafdéd with sevenpy
dollars (to be shared equally among the
members). Therefore; ply to do your best

so that your group's chances of bettering
the other groups and receiVIng the reward
willee increased. YOu will all have ten

 

 

50

minutes to work on each task. I will tell
you when the first ten minutes are up. At
that time you will stOp work and I will
hand out the second task you are to work
on. Before you begin work on each task,
be sure to read the instructions for each
task carefully. Are there any questions?
........... Do not ask any questions after
we begin. Just use your own best judgment
if you have a question. Also, do not talk
to anyone while you work on the tasks.

After these instructions were read, the PIA was handed
out face down to each subject by the eXperimenter. Once
every subject had the PIA, the experimenter then told the
subjects to turn it over and to read the instructions to
the task silently while he read them aloud. Following ten
minutes of work on the PIA, the subjects were told to step
and the PLA was then handed out by the experimenter. The
same procedure used in administering the PIA was used for
the PLA.

Upon completing work on both tasks, all the subjects
were thanked for participating and were informed that the
winning group would be contacted and given the seventy
dollar prize (ten dollars for each group member) within
one week. Each subject was then given two dollars as he
or she left the classroom.

In all conditions subjects worked on the tasks in-
dependently. Also, to insure that the subjects in all
conditions were given exactly ten minutes to work on

each task, the experimenters used stop-watches to keep

track of time.

51

Paranthetically, it is worthwhile to note that the
administration of the experiment went very smoothly. NO
problems at all arose in any treatment condition. Indeed,
no questions were asked in any condition, thus suggesting
that the subjects clearly understood what they were expected

to do during each phase of the experiment.

Scoring

Soon after all testing was completed, each test
paper was given a number which was used as a code for the
name, sex, and treatment condition of each subject. All
the tests were then randomly shuffled together. This was
done in order to avoid experimenter bias in the scoring
of the tests, i.e., to insure that the scorer did not
know the sex or treatment condition to which each subject
belonged.

Although the scoring guide provided by Torrance helps
make scoring as objective as possible, there is typically
some degree of subjectivity in scoring since not all
possible re5ponses are covered by the manual. Therefore,
to check on the reliability of the scores, two raters
(the principal investigator and another graduate student)
independently scored the responses of each subject.
Subsequent Pearson product moment correlations between
the scores of the two raters for each dependent variable

yielded the following inter-rater reliability coefficients:

52

Verbal fluency: .99 ' Figural fluency: .99

Verbal flexibility: .97 Figural flexibility: .98
Verbal originality: .95 Figural originality: .97
Verbal elaboration: .94 Figural elaboration: .97

Such high reliability coefficients are consistent with
previous research which also found inter-rater reliabilities
on the Torrance tests in the .90's for each measure of
creativity (Halpin 8 Halpin, 1974; Torrance, 1966).

Since the summed scores of the two raters were used
in data analysis, the actual scorer reliability coefficients
for each dependent variable, computed by the Spearman-

Brown PrOphecy formula (Ebel, 1972), were as follows:

Verbal fluency: .99 Figural fluency: .99

Verbal flexibility: .98 Figural flexibility: .99
Verbal originality: .97 Figural originality: .98
Verbal elaboration: .97 Figural elaboration: .98

Data Analysis

 

To analyze the effects Of sex and treatment conditions
and their interaction on the scores obtained on the verbal
and figural creativity tasks, two separate 2 by 3 fixed
effects multivariate analyses of variance (MANOVA's) were
the major statistical procedures employed. One MANOVA was
performed on the four measures of verbal creativity. The
other MANOVA was performed on the four measures of figural
creativity. By using the MANOVA statistical approach,
multivariate F-ratios were computed which enabled the

investigator to test for main effects and interaction effects

53

on the dependent variables when such variables were
considered together or simultaneously. The MANOVA approach
also provided univariate F-ratios for each separate depen-
dent variable. When significant univariate F-ratios for

a treatment effect were Obtained, Tukey pair-wise compari-
sons were performed to determine where the specific treat-
ment differences lay, i.e., to determine which treatments
were significantly different from each other on the par-
ticular dependent variable.

To derive additional information from the data
obtained in this study, two supplementary statistical
procedures were also carried out. First, to determine
the intercorrelations of the eight dependent variables,
correlation matrices both within and across all treatment
conditions were computed. Second, to determine the effect
of the different treatments on verbal originality and
verbal elaboration when verbal fluency is held constant,
analyses of covariance (ANCOVA's) were conducted. (The
reason for doing these particular ANCOVA's is more readily
explained after the results pertinent to the hypotheses
of interest are presented in Chapter III).

All the statistical analyses in this study were
calculated in the computer center at Michigan State Univer-
sity. For all analyses an alpha level of .05 was adOpted
to test the statistical significance of the results. The

results are reported in Chapter III.

54

Limitations

 

The generalizability of the results of this study is
limited by the particular methods that were employed.
Specifically, the methodological limitations of this
investigation were:

(1) Subjects were paid volunteer students solicited from
Michigan State University. Therefore, the findings
obtained are generalizable only to comparable
populations.

(2) The verbal creativity task utilized was Torrance's
Product Improvement Activity. The figural creativity
task utilized was Torrance's Parallel Lines Activity.
Therefore, the findings obtained are generalizable
only to creativity tasks that are similar in nature
to these tasks.

(3) Individual competition was induced through promise
of a reward for the individual achieving the best
performance within a group. Group competition was
induced through promise of a reward for the group
achieving the best performance among other groups.
Therefore, the findings obtained are generalizable
only to situations in which individual and group
competition are induced by promise of a reward
for best performance.

(4) In this study, fluency, flexibility, originality,
and elaboration are considered to be important

components of creative thinking. These four traits,

 

55

as expressed in both verbal and figural form, were
the dependent variables measured. The conclusions
drawn from this study, therefore, are only general-

izable to these particular indices of creativity.

CHAPTER III

RESULTS

This chapter presents the results of the statistical
analyses of the data. First, the results of the analyses
relevant to each major hypothesis, and their accompanying
sub-hypotheses, are reported. Next, the results of the
analyses that provide additional or supplementary informa-
tion are presented. Finally, the chapter concludes with
a summary of the findings obtained.

All of the statistical analyses reported in this
chapter were calculated on the Control Data Corporation
6500 Computer System in the computer center at Michigan

State University.

Hypothesis One

 

Hypothesis one stated that verbal creativity is not
differentially affected by non-competitive, individual
competitive, and group competitive situations. A multi-
variate test of this hypothesis was performed and is
displayed in Table 2. The multivariate F-ratio of 2.1644

with D.F. of 8 and 150 has a significant p value of less

56

57

than .0105. Hypothesis one is therefore rejected. The
treatments did differentially influence overall performance
on the verbal creativity task. Inspection of the uni-
variate E-ratios also presented in Table 2 indicates that
the significant multivariate F was primarily caused by the
fact that the treatments differed significantly on the
measures of verbal fluency (p less than .0205), verbal
originality (p less than .0015), and verbal elaboration

(p less than .0023). The sub-hypotheses, which stated that
each Of the above measures is not differentially affected
by the treatment conditions, are therefore rejected also.
However, since the treatment effect on verbal flexibility
only approached significance (p less than .0575), the
correSponding sub-hypothesis for verbal flexibility is not
rejected.

The means and standard deviations for each of the
four measures of verbal creativity that were observed for
each treatment condition are presented in Table 3. It
can be seen that for all measures the scores in the in-
dividual competitive condition were higher on the average
than the scores obtained in the non-competitive or group
competitive conditions. To determine which mean differ-
ences among the treatments were statistically significant,

Tukey pair-wise comparisons were performed on each of the

58

Table 2. Multivariate and Univariate Tests for Treatment
Effects on measures of Verbal Creativity

 

 

Multivariate

 

 

 

 

E-Ratio = 2.1644 2 less than .0105 D.F. = 8, 150 I
Univariate
Variable MS (Error) MS (Hypothesis) FfRatio p value
VF 103.9826 425.5119 4.0921 .0205
VX 16.0979 47.7262 2.9647 .0575
V0 106.7271 755.3929 7.0778 .0015
VE 29.5924 195.1905 6.5959 .0023
D.F. for Hypothesis = 2
D.F. for Error = 78

 

Note: VF=Verbal Fluency, VX=Verbal Flexibility, VO=Verbal

Originality, VE=Verbal Elaboration.

Table 3. Observed means and standard deviations of verbal
creativity measures within each treatment condition

 

 

 

Treatment ylgriable
(n=28) VF VX VO VE
Non-Comp Y 12.9 7.1 7.9 4.7
SD 3.2 1.8 4.0 2.4
Ind Comp 7 16.8 8.4 13.1 6.9
SD 6.3 2.5 6.6 3.3
Grp Comp 7 15.0 7.6 10.0 4.6
SD 5.1 1.6 4.5 2.3

 

Note: The means in this and subsequent tables were computed by

dividing the averages of the summed scores by 2.

Simi-

larly, the standard deviations were computed by dividing
the standard deviations of the summed scores by 2.

59

measures for which significant univariate Esratios were
obtained. Through such post-hoc analysis, it was revealed
that subjects in the individual competitive condition
scored significantly higher than subjects in the non-com-
petitive condition on verbal fluency (p less than .025)
and on verbal originality (p less than .005) and signifi-
cantly higher than subjects in both the non-competitive
and the group competitive conditions on verbal elaboration
(p less than .01 for each). On none of the measures did
the scores of the subjects in the group competitive con-
dition differ significantly from the scores of the subjects
in the non-competitive condition.

Table 3 also indicates that the standard deviations
of the scores for all of the measures of verbal creativity
are highest for the individual competitive condition.
Visual inspection of the raw scores (see appendix B)
suggests that the greater variability of the scores in
the individual competitive condition was attributable to
the fact that more subjects in this condition obtained
extremely high scores on each of the measures than did

subjects in the other two conditions.

Hypothesis Two

 

Hypothesis two was concerned with the question of a

sex by treatment interaction effect on verbal creativity.

60

Formally, it stated that the effects of nonacompetitive,
individual competitive, and group competitive situations
on verbal creativity do not vary with sex. The results
of the multivariate test of this hypothesis are reported
in Table 4. The multivariate E ratio of .5161 with D.F.
of 8 and 150 has a nonnsignificant p value of less than
.843. Hypothesis two is therefore not rejected. From
examination of the univariate F-ratio also presented
in Table 4, it is additionally evident that nothing in
the data supports an interaction effect. Thus, the sub-
hypotheses, which stated that the effects of the treat-
ment conditions on each separate measure of verbal crea-
tivity do not vary with sex, are also not rejected.
Table 5 provides visual corroboration for the lack

of an interaction effect. It can be seen that for both

 

§§§g§ the mean performance on all the measures was highest
in the individual competitive condition and lowest in the
non-competitive condition (the only exception being that
the mean verbal elaboration score for males was negligibly
lower in the group competitive condition than in the non-
competitive condition). A further Observation of interest
is that, with the exception of verbal flexibility, the
mean differences on all of the measures between the in-

dividual competitive and non-competitive and between the

61

 

 

 

 

 

 

Table 4. Multivariate and Univariate Test for Sex Treat-
ment Interaction Effects on measures of Verbal
Creativity

Multivariate
F-Ratio = .5161 p less than .843 D.F 8, 150
Univariate
Variable MS (Error) MS (Hypothesis) F-Ratio p value
VF 103.9826 21.4643 .2064 .8140
VX 16.0979 2.5119 .1560 .8558
VO 106.7271 118.6548 1.1118 .3342
VE 29.5924 13.9048 .4699 .6269
D.F. for Hypothesis = 2
D.F. for Error = 78

 

Table 5. Observed sex by treatment means

Verbal Creativity

for measures of

 

 

 

 

Treatment Variable
(n=14)

VF VX V0 VEL
Non-Comp M 13.1 7.3 8.4 4.8
F 12.6 6.9 7.4 4.5
Ind Comp M 16.5 8.8 11.6 6.4
F 17.0 7.9 14.5 7.4
Grp Comp M 15.6 8.0 10.0 4.6
F 14.3 7.1 9.9 4.5

 

Note:

M= Male, F: Female

62

individual competitive and the group competitive con-
ditions was slightly greater for females than for males.
Indeed, the individual competitive condition seemed to be
especially beneficial for females with respect to the
verbal originality measure. The mean verbal originality
score among females in the individual competitive condition
was nearly two times greater than the mean score among

females in the non-competitive condition.

Hypothesis Three

 

Hypothesis three stated that figural creativity is
not differentially affected.by non-competitive, individual
competitive, and group competitive situations. The
results of the multivariate test of this hypothesis
appears in Table 6. The multivariate E-ratio of 1.6417
with D.F. of 8 and 150 has a non-significant pfvalue of
less than .1176. Hypothesis three is therefore not re-
jected. From inspection of the univariate E-ratios in
Table 6, it appears that one of the figural creativity
measures, figural elaboration, is significantlyaffected
by the treatments (p less than .0173). However, according
to the MANOVA statistical model (Finn, 1974), when-
ever the multivariate F-ratio is found to be non-signifi-
cant, any "significant" univariate F's are interpreted

as being due to chance and are not considered true effects.

Table 6

63

. Multivariate and Univariate Tests for Treatment
Effects on measures Of Figural Creativity

 

 

 

 

 

 

 

 

 

 

Multivariate
F-Ratio = 1.6417 p less than .1176 D.F. = 8, 150
Univariate
Variable MS (Error) MS (Hypothesis) ‘E-Ratio p value
FF 135.3388 73.3690 .5421 .5837
FX 65.6941 69.4405 1.0570 .3525
F0 404.2445 586.3333 1.4504 .2408
FE 408.7866 1749.7262 4.2803 .0173
D.F. for Hypothesis = 2
D.F. for Error ‘ 78
Note: FF = Figural fluency, FX = Figural flexibility, PC = Fig-
ural originality, FE = Figural elaboration
Table 7. Observed means and standard deviations of figural
creativity measures within each treatment condition
Treatment Variable
(“‘23) FF FX F0 FE
Non-Comp Y 11.9 9.8 17.5 15.8
SD 4.9 3.9 8.6 8.5
Ind Comp 7 13.4 10.8 22.1 23.5
SD 5.6 4.0 10.0 11.8
Grp Comp 7 12.0 9.2 19.8 18.1
SD 7.1 4.4 13.5 9.2

 

64

Consequently, all of the sub-hypotheses, which stated that
each separate measure of figural creativity is not diff-
erentially affected by the treatment conditions, are also
not rejected.

The means and standard deviations for each of the
four measures of figural creativity that were observed
for each treatment condition are presented in Table 7.
Inspection of the means reveals a trend which is parallel
to the results obtained on the verbal creativity measures.
More specifically, it can be seen that for all the measures
of figural creativity, mean performance was highest in
the individual competitive condition in comparison to
the non-competitive or group competitive conditions. It
should be re-emphasized, however, that none of the treatment

differences reached statistical significance.

Hypothesis Four

 

Hypothesis four was concerned with the question of
a sex by treatment interaction effect on figural creativity.
Formally, it stated that the effects of non-competitive,
individual competitive, and group competitive situations
on figural creativity do not vary with sex. The results
of the multivariate test Of this hypothesis are presented
in Table 8. The multivariate F-ratio of 1.11 with D.F. of 8

and 150 has a non-significant p value of less than .359.

65

Hypothesis four is therefore not rejected. From
examination of the univariate E-ratios also given in
Table 8, it is additionally evident that nothing in the
data supports an interaction effect. Therefore, the
sub-hypotheses, which stated that the effects of the
treatment conditions on each separate measure of figural
creativity do not vary with sex, are also not rejected.
Table 9 gives the sex by treatment means for each
measure of figural creativity. Although no significant
interactions were obtained, close inspection of Table 9
suggests some interesting patterns. Whereas the mean
performance of females on each of the measures was highest
in the individual competitive condition, males did
not exhibit as consistent a pattern. In fact, on two of
the measures, figural fluency and figural originality,
males achieved their highest mean performance in the
group competitive condition. Another noteworthy pattern
is that whereas the mean performance of males on each
of the measures was lowest in the non-competitive con—
dition, females, on the average, tended to do least well

in the group competitive condition.

66

Table 8. Multivariate and Univariate Tests for Sex by Treat-
ment Interaction Effects on measures of Figural

 

 

 

 

 

 

Creativity
Multivariate
F-Ratio = 1.1111 p less than .3590 D.F. = 8, 150
Univariate
Variable MS (Error) MS (Hypothesis) F-Ratio p value
FF 135.3388 176.4643 1.3039 .2774
FX 65.6941 51.7976 .7885 .4582
F0 404.2445 1137.3333 2.8135 .0661
FE 408.7866 146.8929 .3593 .6993
D.F. for Hypothesis = 2
D.F. for Error = 78

 

Table 9. Observed sex by treatment means for measures of
Figural Creativity

 

 

Treatment

 

 

(n=14) Var1able

FF FX F0 FE

Non-Comp M 12.4 10.5 20.1 17.0
F 11.4 9.0 14.8 14.6

Ind Comp M 13.7 11.1 23.3 22.5
F 13.0 10.5 20.8 24.5

Grp Comp M 14.5 10.8 27.0 18.7
F 9.4 7.6 12.6 17.4

 

67

Summary of Hypothesis Testing

 

To summarize the results of hypothesis testing, the

status of each major hypothesis and its associated sub-

hypotheses are reported below:

Hypothesis 1: Verbal creativity is not differentially

 

affected by non~competitive, individual
competitive, and group competitive situa-
tions. Status: Rejected.

Sub-hypotheses:

 

(1a)

(1b)

(16)

(1d)

Verbal fluency is not differentially affected by
non-competitive, individual competitive, and
group competitive situations. Status: Rejected.

Verbal flexibility is not differentially affected
by non-competitive, individual competitive, and

group competitive situations. Status: Failed to
reject.

Verbal originality is not differentially affected
by non-competitive, individual competitive, and
group competitive situations. Status: Rejected.

 

Verbal elaboration is not differentially affected
by non-competitive, individual competitive, and
group competitive situations. Status: Rejected.

Hypothesis 2: The effect of non-competitive, individual

 

competitive, and group competitive situations
on verbal creativity does not vary with sex.
Status: Failed to reject.

 

Sub-hypotheses:

 

(2a)

(2b)

The effect of non-competitive, individual compet-
itive, and group competitive situations on verbal
fluency does not vary with sex. Status: Failed
to reject.

 

The effect of non-competitive, individual compet-
itive and group competitive situations on verbal
flexibility does not vary with sex. Status:
Failed to reject.

 

(2C)

(2d)

68

The effect of nonscompetitive, individual compet-
itive, and group competitive situations on verbal
originality does not vary with sex. Status:
Failed to reject.

 

The effect of non-competitive, individual compet-
itive, and group competitive situations on verbal
elaboration does not vary with sex. Status:
Failed to reject.

 

Hypothesis 3: Figural creativity is not differentially

affected by non~competitive, individual
competitive, and group competitive situations.
Status: Failed to reject.

 

Sub-hypotheses:

 

(3a)

(3b)

(3C)

(3d)

Figural fluency is not differentially affected by
non-competitive, individual competitive, and group
competitive situations. Status: Failed to reject.

 

Figural flexibility is not differentially affected
by non-competitive, individual competitive, and
group competitive situations. Status: Failed to

reject.

Figural originality is not differentially affected
by non-competitive, individual competitive, and
group competitive situations. Status: Failed to
reject.

Figural elaboration is not differentially affected
by non-competitive, individual competitive, and
group competitive situations. Status: Failed to

reject.

 

 

 

Hypothesis 4: The effect of non—competitive, individual

 

competitive, and group competitive situations
on figural creativity does not vary with sex.
Status: Failed to reject.

 

Sub-hypotheses:

 

(4a)

(4b)

The effect of non-competitive, individual competi-
tive, and group competitive situations on figural
fluency does not vary with sex. Status: Failed to

 

reject.

The effect of non-competitive, individual compet-
itive, and group competitive situations on figural
flexibility does not vary with sex. Status:
Failed to reject.

 

69

(4c) The effect of non-competitive, individual compet-
itive, and group competitive situations on figural
originality does not vary with sex. Status:
Failed to reject.

 

(4d) The effect of non-competitive, individual compet-
itive, and group competitive situations on figural
elaboration does not vary with sex. Status:
Failed to reject.

 

Additional Analyses

 

After carrying out the statistical analyses that were
necessary to test the formal hypotheses of this study, the
investigator believed that additional data analysis would
be valuable in that it could provide supplementary infor-
mation that was of interest. Consequently, further statisti-
cal analyses were performed with the objectives of answering
the following questions: (1) After collapsing the data
across treatments, were any significant sex differences
obtained with respect to performance on the verbal and
figural creativity tasks? (2) To what extent did the measures
of verbal and figural creativity correlate with each other
both within and across all treatment conditions? and (3) Was
there still a treatment effect on the measures of verbal
originality and verbal elaboration when verbal fluency was
held constant? The results that were obtained with respect

to these questions are reported below in turn.

Effect of Sex on Verbal Creativipy
The MANOVA that was used to test for a treatment effect
and a sex by treatment interaction effect on the verbal

creativity scores also, of course, provided F-ratios for the

7O

testing of a sex main effect. The results of this multi-
variate test are presented in Table 10. As can be seen,
both the multivariate F (§= 1.25, p less than .295) and

all the univariate F's are non-significant. Therefore, no
significant sex effect was obtained, i.e., collapsing
across treatments, males and females did not differ signif-
icantly in their performance on the verbal creativity task.
Table 11, which gives the means of the males and females on
each of the verbal creativity measures, visually confirms

this finding.

Effect of Sex on Figural Creativity

 

Like the MANOVA on the verbal creativity scores, the
MANOVA that was used to test for a treatment main effect
and a sex by treatment interaction effect on the figural
creativity scores also provided F ratios for the testing
of a sex main effect. The results of this multivariate
test are reported in Table 12. The mult1variate Efratio
of 4.9096 with D.F. of 4 and 75 has a highly s1gn1ficant p
value of less than .0015. This indicates that, collapsing
across treatments, the sexes differed significantly in
terms of their overall performance on the figural creativity
task. Inspection of the univariate Feratios shows that
the significant multivariate F was primarily caused by the
fact that the sexes differed significantly on the measures
of figural f1ex1bility (p less than .0442) and figural

originality (p less than .0011).

71

Table 10. Multivariate and Univariate Tests for a Sex Effect
on measures of Verbal Creativity

 

 

Multivariate

 

F-Ratio = 1.2557 p less than .2950 D.F. 75

ll
h
w

 

Univariate

 

Variable MS (Error) MS (Hypothesis) ‘F-Ratio p value

 

VF 103.9826 18.1071 .1741 .6777
VX 16.0979 41.4405 2.5743 .1127
V0 106.7271 29.7619 .2789 .5990
VE 29.5924 4.2976 .1452 .7042
D.F. for Hypothesis = 1
D.F. for Error = 78

 

Table 11. Observed means of males and females for measures
of Verbal Creativity

 

 

 

 

Sex .
(n=42) Variable .

VF VX V0 VE
Males 15.1 8.0 10.0 5.3
Females 14.6 7.3 10.6 5.5

 

Table 13, which presents the means for males and
females on each figural creativity measure, reveals that

the males did better than the females on the figural task.

72

Table 12. Multivariate and Univariate Tests for a Sex
Effect on measures of Figural Creativity

 

 

Multivariate

 

E-Ratio = 4.9096 p less than .0015 D.F. = 4, 75

 

Univariate

 

Variable MS (Error) MS (Hypotheses) FeRatiO p value

 

FF 135.3388 438.8571 3.2427 .0757
FX 65.6941 275.0476 4.1868 .0442
F0 404.2445 4710.0119 11.6514 .0011
FE 408.7866 30.9643 .0757 .7839

D.F. for Hypothesis = l
D.F. for Error = 78

 

Table 13. Observed means of males and females for measures
of Figural Creativity

 

 

Sex

 

 

(n-42) Variable
FF FX F0 FE
Males 13.5 10.8 23.5 19.4

Females 11.3 9.0 16.1 18.8

 

73

On all the measures males outperformed females, although
only in the case Of figural flexibility and figural orig-

inality were these differences considered significant.

Intercorrelations of verbal and figural creativity measures

 

To determine the extent of the inter-relationships
among the measures of verbal and figural creativity across
all treatment conditions, Pearson product moment correla-
tions between each measure and every other measure were
calculated. The correlation matrix that resulted is pre-
sented in Table 14.

Examination of the matrix reveals several noteworthy
findings. First, virtually all of the intercorrelations
were statistically significant--the only exceptions being
verbal elaboration with figural fluency (p less than .11)
and verbal elaboration with figural flexibility (p less
than .13). Indeed, while the average intercorrelation
among all the measures is .41, several intercorrelations
were extremely high, reaching above an intercorrelation of
.70. A second finding worthy of note is that while the
average intercorrelation among the verbal creativity measures
was a moderately high .51, the average intercorrelation
among the figural creativity measures was also moderately
high, being .62. Interestingly, however, the average
intercorrelation between the verbal and figural creativity
measures was only .29. These findings show that the

verbal measures correlated more highly among themselves

74

than they did with the figural measures. Similarly, the
figural measures were more related to each other than they
were with the verbal measures.

To determine if the degree of the inter-relationships
among the verbal and figural creativity measures was differ-
ent under the different treatment conditions, correla-
tion matrices between the scores within each treatment
condition were also computed. Table 15 presents the corre-
lation matrix for the non-competitive condition, Table 16
presents the correlation matrix for the individual compet-
itive condition, and Table 17 presents the correlation
matrix for the group competitive condition. Close in-
Spection of these matrices reveals that 9 of the 28 inter-
correlations in the non-competitive condition were sta-
tistically signigicantly, 19 of the 28 intercorrelations
in the individual competitive condition were statistically
significant, and 16 of the 28 intercorrelations in the
group competitive condition were statistically signifi-
cant. These findings indicate that the measures of ver-
bal and figural creativity were most related to each other
in the individual competitive condition and least related
to each other in the non-competitive condition. The im-

plications of these results are presented in Chapter IV.

75

Table 14. Intercorrelations of Verbal and Figural Creativity
Measures across all Treatment Conditions (N = 84)

 

 

Measure vx v0 ve FF FX F0 FE
VF .76a .71a .33a .34a .35a .37a
vx .53a .22C .24C .26b .32b
v0 .38a .34a .39a .38a
VE .14 .13 .24C .zeb
FF .93a .84a .32b
FX .803 .35a
F0 .483

 

:Significant at .001 level; bSignificant at .01 level;
Signifcant at .05 level

Table 15. Intercorrelations of Verbal and Figural Creativity
Measures within Non-Competitive Condition (N = 28)

 

 

Measure VX VO VE FF FX F0 FE
VF .68a .49b 15 31 25 22 24
vx .44b .11 .23 .10 .24 .33c
v0 .553 .15 .01 .01 .13
VE -.17 -.23 -.18 -.01
FF .90a .78a .08
FX .73a .07

F0 .44

 

3Significant at .001 level; bSignificant at .01 level;
Significant at .05 level

76

 

 

Table 16. Intercorrelations of Verbal and Figural Creativity
Measures within Individual Competitive Condition
(N = 28)
Measure VX V0 VE FF FX F0 FE
VF .81a .763 .36C .36C .46b .38C .30
vx .57a .15 .29 .39C .31 .27
V0 42C 34C .36C 43C 39C
VE .01 .06 .25 .26
FF .94a .81a .45b
FX .788 .49b
F0 s7a

 

a Significant at .001 level; b Significant at .01 level;

C

Significant at .05 level

 

 

Table 17. Intercorrelations of Verbal and Figural Creativity
Measures within Group Competitive Condition (N = 28)

Measure VX V0 VE FF FX F0 FE
VF .69a .64a .29 .31 .25 .32C .35c

VX .31 .17 .10 .10 .17 .15

V0 .34C .57a .54a .52b .26
VE .42c .41c .43C .17

PF 953 .89a 30
FX 878 .38c
F0 41C

 

aSignificant at .001 level;vbSignificant at .01 level;

CSignificant at .05 level

77

Covariate Analysis: Verbal originality and Verbal elaboration

From Table 14 it is evident that verbal fluency has
a significant high correlation with verbal originality
(.71, p less than .001) and a significant moderate corre-
lation with verbal elaboration (.36, p less than .001).
These positive correlations indicate, of course, that as
verbal fluency increases, these other two variables also
tend to increase. In light of this fact, an interesting
question that arose was whether or not the significant
treatment effect that was obtained on verbal originality
and verbal elaboration would still be maintained if the
effect of verbal fluency on these measures was accounted
for, i.e., held constant. To answer this question, analyses
of covariance (ANCOVA's) on the verbal originality
scores and on the verbal elaboration scores with verbal
fluency as the covariate were carried out. The results of
these ANCOVA's are presented in Tables 18 and 19.

As can be seen from Table 18, when verbal fluency was
held constant, no longer was there a significant treatment
effect on verbal originality, although it did approach
significance (p less than .0587). While this finding does
not nullify the previously reported finding that the in-
dividual competitive condition was significantly better than
the non-competitive condition for enhancing verbal original-
ity, it does suggest that the significant increase of verbal
originality in the individual competitive condition was

largely a secondary result of the increase of verbal

78

fluency in that condition.

The results of the ANCOVA on verbal elaboration,
however, do not lead to the same conclusion. Table 19
shows that the effect of the treatment on verbal elaboration
was still significant (p less than .0136) even when verbal
fluency was held constant. This indicates that the signifi-
cant increase in verbal elaboration in the individual
competitive condition relative to the non-competitive and
group competitive conditions was not a secondary result

of the increase in verbal fluency.

Table 18. Analysis of Covariance: Verbal Originality

 

 

 

Sources of Variation df MS F
Treatment 2 169.81 2.94*
Sex ' 1 . 70.39 1.22
Treatment by Sex 2 74.50 1.29
Within 77 57,71

 

*p less than .0587, not significant at .05 level.

Table 19. Analysis of Covariance: Verbal Elaboration

 

 

 

Sources of Variation df MS F
Treatment 2 124.58 4.55*
Sex 1 7.51 .27
Treatment by Sex 2 9.89 .36
Within 77 27.37

 

*p less than .0136, significant at .05 level.

(1)

(2)

(3)

79

Summary of Findipgs

 

To summarize, the major findings of this study were:
Verbal creativity (as measured by verbal fluency,
verbal flexibility, verbal originality, and verbal
elaboration) was differentially affected by non-
competitive, individual competitive, and group compet-
itive situations. Specifically, subjects in the
individual competitive condition did significantly better
than subjects in the non-competitive condition on verbal
fluency and verbal originality and significantly better
than subjects in both the non-competitive and group
competitive conditions on verbal elaboration. On none
of the measures did the subjects in the non-competitive
condition differ significantly from subjects in the
group competitive condition.

Figural creativity (as measured by figural fluency,
figural flexibility, figural originality, and figural
elaboration) was not differentially affected by non-
competitive, individual competitive, and group compet-
itive situations. Although average performance on

all measures of figural creativity was generally
highest in the individual competitive condition and
lowest in the non—competitive condition, none of these
treatment differences were significant.

The effects of non—competitive, individual competitive,
and group competitive situations on verbal creativity

did not vary with sex, i.e., no sex by treatment

(4)

(5)

(6)

8O

interaction occurred. The Observed cell means on each
measure of verbal creativity indicated that fp:_pp£h
gegeg, average performance was highest in the individual
competitive condition and lowest in the non-competitive
condition.

The effects of non-competitive, individual competitive,
and group competitive situations on figural creativity
did not vary with sex, i.e., no sex by treatment inter-
action occurred. The observed cell means on each
measure of figural creativity suggested some interesting
patterns, however. Among females, average performance
was highest in the individual competitive condition

and generally lowest in the group competitive condition.
Among males, on the other hand, average performance

was highest in either the individual competitive or
group competitive conditions and lowest in the
non-competitive condition.

Collapsing across treatments, males and females did

not differ significantly on the measures of verbal
creativity. Significant sex differences were obtained
on figural creativity, however. Specifically, males
did significantly better than females on figural
flexibility and figural originality.

Almost all of the measures of verbal and figural
creativity were significantly related to one another.
However, verbal creativity measures correlated more

highly among themselves than they did with the figural

(7)

(8)

81

measures. Similarly, the figural measures correlated
more highly among themselves than they did with the
verbal measures.

The degree of relationship among the verbal and
figural creativity measures varied under different
treatment conditions. Highest intercorrelations were
obtained in the individual competitive condition and
lowest intercorrelations were obtained in the non-
competitive condition.

With verbal fluency held constant, no longer was there
a significant treatment effect on verbal originality.
This was not the case for verbal elaboration, however.
Even when verbal fluency was held constant, verbal
elaboration was still significantly affected by the

treatments .

CHAPTER IV

DISCUSSION
This chapter presents a discussion of the major
findings of the study. In addition, some suggestions for

future research are offered.

Discussion of Findipgg
To facilitate organization of the discussion, the
major findings of this study are discussed under the
following headings: (1) Treatment effect on verbal
creativity; (2) Treatment effect on figural creativity;
(3) Sex by Treatment interaction effects; (4) Sex differ-
ences in verbal and figural creativity; and (5) Relation-

ships among verbal and figural creativity measures.

Treatment effect on verbal creativity

Previous research indicated that individual competi-
tion, in which a reward was offered for best performance,
can enhance the performance of elementary school students
(Torrance, 1959, 1965) and junior high school students
(Raina, 1968) on a verbal creativity task. The present
investigation extends this finding to college students.
In comparison to the subjects in the non-competitive
situation, subjects in the individual competitive situation

scored significantly higher on the measures of verbal

82

83

fluency, verbal elaboration, and verbal originality. In
more descriptive terms, subjects in individual competition
generated a significantly greater number of ideas that
were both meaningful and relevant to the task at hand.
They also, to a significantly greater extent, embellished
or added detail to their ideas so that they were more
interesting or complex. Perhaps most importantly, subjects
in individual competition produced significantly more
ideas that were of high quality; ideas that were clever,
original, or uncommon, yet still apprOpriate to the task.
Although the covariate analysis suggested that this latter
finding was largely a secondary result of the positive
effects individual competition had on verbal fluency, this
still does not negate the fact that verbal originality,
when measured according to Torrance's scoring procedures,
was significantly enhanced by individual competition. The
results of the covariate analysis do suggest, however, that
if the number of ideas in response to the task was held
constant across all subjects (e.g., if all subjects were
told to produce five ideas apiece), individual competition
may not have significantly benefited verbal originality
production. Further research is needed to verify this
possibility.

In contrast to the stimulating effects of individual
competition, it was found that group competition did not
effectively promote performance on the verbal creativity
task. Indeed, not only were the scores of subjects in

the group competitive situation not significantly different

84

from the scores of subjects in the non—competitive situa-
tion, but also, subjects in group competition produced
significantly less verbal elaboration responses than the
subjects in individual competition. A possible explanation
for the failure of group competition to stimulate creative
performance may be derived from considering the nature
of the group competitive situation itself. Recall that
in the group competitive condition the subjects were
informed that they would receive the reward of seventy
dollars (to be shared equally among them) if the overall
performance of their group was better than that of the
other groups. This means, of course, that each individual
was largely dependent upon his teammates in order to
receive the reward. For even if an individual did
exceptionally well on the tasks, he still may not win the
reward unless his teammates also did well. In essence,
individuals in group competition had less "personal control"
over the possibility of winning the reward than did indi-
viduals in individual competition. To the extent that the
subjects in the group competitive situation were cognizant
of this fact, it may have lessened their motivation to do
well, and, as a consequence, their performance was no
better than the performance of subjects in the non-
competitive situation, in which no incentives were offered.
Regardless of whether or not this explanation is
correct, it is clear that group competition was not as
effective as individual competition. This finding gives

support to a very important conclusion: the extent to

85

competition enhances performance on a verbal creativity
task may well depend upon the type of competition that

is utilized. This conclusion, it may be remembered, is in
accord with much of the research on non-creative tasks
(e.g., Maller, 1929; Julian 8 Perry, 1967), in which
individual competition and group competition were found

to have differential effects.

Treatment effect on figural creativity

 

In addition to assessing the effect of non-competitive,
individual competitive, and group competitive situations
on verbal creativity, this study also sought to determine
what effect these three conditions would have on figural
creativity. Previous research (Brown, 1973) had suggested
that although a particular procedure may be useful for
stimulating verbal creativity, it may not be beneficial
to figural creativity. The results of this study are
substantially in agreement with this particular notion.
Although subjects in the individual competitive condition
tended to do slightly better on the average than subjects
in the non-competitive and group competitive conditions
on all of the measures of figural creativity, none of
these average differences were found to be statistically
significant. One must assume, therefore, that the differences
were simply due to chance.

It is curious to this investigator that performance on
the figural creativity task was less responsive to the

treatment conditions than was performance on the verbal

86

creativity task. There does not appear to be any Obvious
explanation as to why this result occurred. A plausible
explanation, though speculative, however, is suggested by
the work of Clifford and her associates (Clifford, 1972;
Clifford, Cleary, and Walster, 1972). Based on a number of
studies, these investigators have come to the conclusion
that although competition can be an effective stimulator of
performance on a variety of tasks, it has less impact on
tasks which are "intrinsically interesting". They theorize
that when an individual perceives a task as intrinsically
interesting or challenging, he is highly motivated to work
and perform on this task; therefore, the imposition of
extrinsic motivational techniques, such as competition, are
less influential simply because they are unnecessary to
establish and promote motivation to perform. Kruglanski
and his associates (1975) have provided some empirical
support for this theory. These investigators found that
college students were significantly more willing to work

on a task without monetary compensation when they perceived
it to be an interesting one than when they considered the
task to be dull.

With regard to the results of this investigation, is
it possible that the figural creativity task was perceived
as more intrinsically interesting than the verbal creativity
task? If so, to the extent that the aforementioned theory
is correct, this could be the reason performance on the

figural creativity task was not significantly enhanced

87

by the competitive treatments. However, since the
investigator did not gather systematic evidence with
respect to the intrinsic interest of each task, this
explanation must remain an interesting speculation that
can be verified only through future empirical research.

In any event, an important conclusion to be noted is that
just as the effectiveness of competition for stimulating
creativity may depend upon the nature of the competitive
situation (i.e., individual vs. group), it also may depend
upon the nature of the creative task (i.e., verbal vs

figural).

Sex by Treatment interaction effects

Another major concern of this study was to test whether
or not the sexes were differentially affected by the
treatment conditions. Inspection of the sex-typing
literature (Mischel, 1970) and research by Richman (1972)
had suggested the possibility that whereas the performance
of males on the creative tasks might be most enhanced by
individual competition, the performance of females might
be greater in group competition. Analysis of the results
of this study provided no support for such an interaction
effect. Most generally, overall performance for both
sexes tended to be highest in the individual competitive
situation. Indeed, in contrast to what might be expected,
individual competition seemed to be especially beneficial
to the females. Not only did females achieve their highest

Imean performance on every single measure of verbal and

88

figural creativity in individual competiton, but also,
the differences in the creativity scores between the
individual competitive condition and the other two conditions
were generally greater among females than among males.
That females tended to do best when they competed
individually against other males and females seems to
contradict the "fear of success" theory postulated by
Horner (1972). According to this theory, it is held that
females in our culture are taught that competition,
particularly with a male, is unfeminine and may result in
social rejection. As a result of such training, Horner
argues that many women develOp a "motive to avoid success"
which serves to inhibit their performance in individual
competitive situations. Indeed, Horner also asserts that
those women who are most capable of achieving success
(e.g., college women) are the ones who tend to be most
afraid of it. Although Horner has marshalled some evidence
to support her theory (Horner, 1968), more recent research
(Zuckerman and Wheeler, 1975) has failed to find any firm
evidence of a "fear of success" motive among women.
Certainly the results of the present investigation offer
no support for Horner's theory. Perhaps the pervasive
growth of the women's liberation movement in recent years

has served to reduce "fear of success" among women.

Sex differences in verbal and figural creativity
Although it was not a primary focus of this

investigation, data with regard to sex differences in verbal

89

and figural creativity was also obtained. After collapsing
the data across treatment conditions, analysis of the
results indicated no sex differences on the measures of
verbal creativity. It was found, however, that males

did significantly better than females on the measures of
figural flexibility and figural originality. These
findings, though interesting, are probably quite limited

in terms of generalizability. Examination of the empirical
literature concerned with sex differences in creativity
reveals that although a number Of studies have Obtained
results largely consistent with those of this study (e.g.,
Bieri, Bradburn, G Galinsky, 1958; Mendelsohn G Griswold,
1966), just as many studies have reported partially or
totally contradictory results (e.g., Torrance 8 Aliotti,
1969; Frederickson 8 Evans, 1974). Indeed, an exhaustive
review of the research in this area by Maccoby and Jacklin
(1974) points up the highly inconsistent and contradictory
nature of this type of research. In light of this fact,
the sex differences observed in this study should be
interpreted with extreme caution. It is probably best to
assume that the differences obtained are largely a function
of the particular sample studied or the particular tasks

employed. Broader generalizations do not seem warranted.

Relationships among verbal and figural creativity measures
A further consideration of this study was the extent
to which the measures of verbal and figural creativity

related to one another. When Pearson product moment

90

correlations were computed among all the measures across all
the treatment conditions, the results showed that almost
all the intercorrelations were statistically significant.
This indicates that each variable was measuring, at least
to some degree, some factor in common with every other
variable. Of course, this finding should be expected if

we are to believe that each measure is an index of the
larger construct of creativity. In this regard, it was
also revealed that whereas the verbal measures tended to
correlate highly among themselves (average intercorrelation
of .51), and the figural measures tended to correlate highly
among themselves (average intercorrelation of .62), the
intercorrelations between the verbal and figural measures
were comparatively low (average intercorrelation of .29).
Two important implications are suggested by this finding.
First, it supports the value of examining creativity along
two separate dimensions, one verbal and the other figural.
A second and related implication is that one should not
assume that verbally creative individuals are also
figurally creative. Similarly, figurally creative
individuals should not be assumed to be verbally creative.
This would seem to be an especially important fact for
teachers to keep in mind.

Additional information with respect to the inter-
relationships among the verbal and figural creativity
measures was provided when intercorrelations were computed
within each separate treatment condition. In general, it

was found that the intercorrelations among the measures were

91

highest in the individual competitve condition and lowest
in the non-competitive condition. This finding suggests
that the degree of the relationships that are obtained
among creativity measures may well depend upon the conditions
under which the tests are administered. It is noteworthy
that several other investigators have Observed this to

be true, although the results they have obtained have

been contradictory. For example, although Leith (1972)
found, similar to this study, that intercorrelations

among creativity measures were highest in an evaluative
rather than a relaxed atmosphere, Vernon (1971) and Boersma
and O'Bryan (1968) reported higher relationships among
creativity measures when the tests were given under

relaxed rather than evaluative conditions. While such
contradictory results are not readily explained, the major
point to be emphasized is that different testing conditions
can differentially influence the extent of the inter-
relationships among measures Of creativity. In light

of this fact, it seems quite possible that one of the
reasons research has been so inconsistent with reSpect to
the reliability and validity of creativity tests, includ-
ing the Torrance tests, is that different investigators

have administered the tests under different conditions.

92

Suggestions for Future Research

At a most general level, the results of this study
support the idea that the level of creative productivity
of an individual can be influenced by the conditions under
which he works. Specifically, the results of this study
indicate that individual competition, in which individuals
compete against other individuals in order to win a concrete
reward, can effectively enhance creative output on a verbal
task. An Obvious educational implication of this finding,
of course, is that teachers may find it profitable to
arrange individual competitive situations in the classroom
when verbal creativity is one of their instructional
objectives. However, before any definitive and broad
implications can be drawn, additional investigations that
overcome the limitations of the present study are needed.

For example, both in this study and in the studies
by Torrance (1959, 1965) and Raina (1968) money or concrete
prizes were used as the reward in the competitive situations.
It would be interesting to examine whether different types
of rewards, such as grades or public recognition, would
lead to outcomes different from those obtained in this study.
Indeed, it is reasonable to assume that a greater amount of
anxiety would be generated in a situation in which individ-
uals competed against each other for grades rather than for
Inoney. And since there is research to suggest that anxiety
21nd.stress may impede creative thinking (Tortorella, 1967;

lCrOp, Alegre, 8 Williams, 1969), it is quite possible that

93

individual competition for grades would have a different
effect on creative performance than individual competition
for money.
Another useful study would be to test the effects of
competition on performance on a more substantial task
than the one used in this study. For example, would
competition facilitate the production of creative research
papers or creative paintings? Clearly working on Torrance's
creative thinking tests is a much simpler task than the
task of producing a research paper or painting. Such
differences in the complexity of the task are certainly a
factor worthy of consideration in future research.
Since this study utilized only independent creative
tasks (i.e., individuals worked separately on the tasks),
a further investigation that might be fruitful would be
to examine the relative effects of non-competition, indi-
vidual competition and group competition on performance
on an interdependent creative task (i.e., a task on which
individuals work together). As reported in Chapter I,
.a number of studies found that group competition was more
laeneficial than individual competition for increasing
Irroductivity on non-creative interdependent tasks. Similar
effects might be predicted on creative interdependent tasks.
Still another study that might be worthwhile to
cc3r1duct would be to assess how individual and/or group
Cc>nnpetition compares to other procedures for stimulating

Cr‘eezitive production. For instance, would individual

94

competition be a significantly better technique for
promoting creativity than brainstorming or modeling?
Only through comparative research can such questions be
answered.

In conclusion, as with other areas of creativity
research, our understanding of the relationship between
competition and creativity is still in its early phases.
There is still much yet to be done. While the results of
this study support the efficacy of individual competition
as a means for enhancing verbal creativity, further research
is needed before more definitive and wide-ranging conclusions

can be Offered.

APPENDICES

APPENDIX A
DEMOGRAPHIC DATA SHEET

95

NAME:

 

SEX: Male ___ Female

AGE:

LEVEL OF COLLEGE:

Freshman
SOphomore
Junior
Senior

Grad

APPENDIX B
RAW SUMMED SCORES

96

Raw Summed Scores for Ss in the Non-competitive Conditions_r

 

 

 

 

Sex Variables

(“‘14) VF vx vo VE FF FX F0 FE
20 16 16 14 24 24 48 25
26 12 1o 7 24 20 52 28
31 16 33 14 8 8 14 46
20 12 6 6 18 18 34 4o
38 18 35 13 36 34 52 37
24 14 13 4 52 40 69 33

Males 30 20 25 9 32 18 68 77
20 10 11 17 22 20 36 34
27 18 14 8 12 12 11 22
32 16 18 31 28 38 35
36 19 16 14 24 18 40 34
14 8 15 7 14 12 17 22
20 13 8 9 28 24 48 8
29 12 14 8 23 19 35 36
18 12 16 9 16 12 30 33
26 14 12 8 30 20 28 12
28 16 18 16 16 16 3o 24
21 12 12 1 34 30 44 19
32 18 33 17 32 14 20 8
26 9 16 12 26 24 36 25
24 8 6 6 12 12 2 10
28 17 18 16 24 20 20 40

Females 36 16 12 4 30 24 44 20
22 10 18 16 16 12 26 50
20 13 22 10 30 24 42 30
34 19 8 3 30 26 61 80
14 12 2 4 14 12 26 36
24 18 14 5 10 6 6 22

 

97

Raw Summed Scores for Ss in Individual Competitive Condition

 

 

 

Sex Variables

(n=14) VF VX VO VE FF FX FO FE
23 15 23 15 16 16 38 62
38 18 31 23 37 30 87 54
50 29 39 26 36 32 65 62
30 14 26 15 21 18 36 33
34 21 l6 13 20 20 37 52
50 25 37 8 l6 16 36 41'

Males 20 16 8 8 18 18 38 65
12 8 2 7 32 20 58 45
27 16 14 17 16 8 36 16
59 26 4S 9 52 41 76 93
40 19 24 16 16 12 24 37
24 12 10 5 32 26 40 3
35 16 31 12 28 25 44 34
21 12 18 S 43 28 37 34
37 18 11 11 34 30 48 44
45 19 45 6 22 19 44 38
60 19 35 23 25 23 37 26
31 15 29 21 6 6 8 22
18 14 13 7 15 12 21 31
14 12 2 3 18 18 24 44
41 23 37 9 42 30 64 38

Females 34 18 44 18 33 26 52 50
34 19 37 11 28 18 38 41
39 14 46 23 44 30 100 106
46 20 34 19 36 28 46 106
22 8 28 18 18 18 28 50
28 12 27 17 14 12 25 52
26 12 18 22 29 23 46 38

 

 

Raw Summed Scores for S5 in Group Competitive Condition

98

 

 

 

 

Sex Variables

(n-l4) VF VX VO VE FF FX FO FE
24 12 6 5 24 20 40 38
42 21 32 10 38 30 62 46
22 16 8 9 14 14 34 22
13 13 8 2 16 14 36 47
26 16 26 22 21 48 23
29 13 14 15 36 28 66 69

Males 20 14 12 16 14 30 22
28 18 14 12 8 25 9
39 19 29 10 58 38 128 66
38 14 25 1 15 13 28 28
35 12 41 17 70 42 106 12
56 22 34 10 40 28 62 78
36 20 12 18 l6 14 44 28
30 14 20 10 30 20 54 38
26 14 28 10 15 15 23 25
44 20 23 14 20 18 24 33
29 10 21 10 l4 13 32 46
38 18 28 4 10 8 18 20
25 14 20 8 30 28 24 40
40 15 25 17 18 14 42 42

Females 32 16 20 12 14 12 18 48
36 15 6 1 12 6 0 8
16 12 14 5 10 10 14 38
40 14 29 10 24 16 42 72
22 16 17 6 22 18 24 34
19 12 16 8 17 13 22 31
18 13 14 13 29 16 22 16
16 10 15 9 27 25 47 35

 

REFERENCES

REFERENCES

Adams, J. ”The relative effects of various testing atmos-
pheres on spontaneous flexibility, a factor of
divergent thinking." Journal of Creative Behavior,
3, 1968, 187-194.

 

Adler, A. Social interest: A challenge to mankind (1933).
New York: Capricorn Books, 1964:

Ausubel, D.F. Educational Psycholo : A cognitive View.
New York: Holt, Rinehart, Winston, 1968.

Bamber, R., Jose, P., 8 Boice, R. "Creativity as affected
by differential reinforcement and test instructions."
Bulletin of the Psychonomic Society, 9, 1975, 361-
363.

Bandura, A. Principles of Behavior Modification. New York:
Holt, Rinehart, 8*Winston, 1969.

Bardwick, J.M. Psychology of Women: A study of biocultural
conflicts. New ork: Harper 6 Row, 1971, pp. 154-
166.

 

Belcher, T. "Modeling original divergent responses: An
initial investigation." Journal of Educational
Psychology, 91» 1975, 351-358.

 

 

Bieri, J., Bradburn, W., a Galinsky, M. "Sex differences
in perceptual behavior." Journal of Personality,
59, 1958, 1-12.

 

Boersma, F. 8 O'Bryan, K. "An investigation of the rela-
tionship between creativity and intelligence under
two conditions of stress." Journal of Personality,
g9, 1968, 341-348.

 

Brown, R. "An investigation of the effects of feedback
and social condition on creativity and self-concept."
Dissertation Abstracts International, 33 (9-A),
1973, 4928-4929.

Brown, L., Hoppler, B., VanDeventer, P., 8 Sontag, E.
Use of the 'college bowl' format to increase silent
reading comprehension." Child Study Journal, i, 1973,
181-193.

99

100

Carment, D. G Hodkin, B. "Coaction and competition in
India and Canada." Journal of Cross-cultural
Psychology, 1, 1973, 459-469.

 

Clifford, M. "Motivational effects of competition and goal
setting in reward and non-reward conditions."
Journal of Experimental Education, g2, 1971, 11-16.

Clifford, M. "Effects of competition as a motivational
technique in the classroom." American Education
Research Journal, 3, 1972, 123-137.

 

 

Clifford, M., Cleary, T.A., & Walster, C.W. "Effects of
emphasizing competition in classroom-testing
procedures." Journal of Educational Research, eg,
1972, 234-238.

Crockenberg, S.B. "Creativity tests: A been or boondazzle
for education." Review of Educational Research,
.11, 1972, 27-46. “

Cronbach, L.J. Educational peychology, New York: Harcourt,
Brace 6 WOrld, 1963T

Dellas, M. & Gaier, E. "Identification of creativity: the
individual." Psychological Bulletin, Zé, 1970,
55-73.

Deutsch, M. "The effects of cooperation and competition
upon the group process." In D. Cartwright G
A. Zander (Eds.), Group Dynamics, Elmsford, N.Y.:
Row, Peterson, 8 Co., 1962.

 

Ebel, R. Essentials of Educational Measurement. Englewood
Cliffs, New Jersey: Prentice-Hall, 1972.

Finn, J.D. A General Model for Multivariate Analysis.
New YOrk: Holt, Rinehart, E Winston,—1974.

 

Fredericksen, N. G Evans, F.R. "Effects of models of
creative performance on ability to formulate hypo-
theses." Journal of Educational Psychology, 99,
1974, 83-89.

Freischlag, J. "A comparison of the effects of sex, compe-
tition, and ability on a perceptual-motor task."
Research Quarterly, 11, 1973, 178-184.

IOl

Gilchrist, M. 8 Taft, R. "Originality on demand."
Psychological Reports, 31, 1972, 579-582.

 

Golann, S. Psychological study of creativity." Psychologi-

 

cal Bulletin, 60, 1963, 548-565.

 

Guilford, J.P. Intelligence, Creativity; and their
educational implications. San Diego: Knapp, 1968.

 

 

Guilford, J.P. The nature of human intelligence. New York:
McGraw-Hill, 1967.

 

Guilford, J.P. "Traits of creativity." In H.H. Anderson
(Ed.), Creativity and its cultivation. New York:
Harper 8 Row, 1959, pp. 142-162.

 

Guilford, J.P. "Creativity." American Psychologist, g,
1950, 444-454.

 

Halpin, G. 8 Halpin, G. "Can self-trained scorers reliably
score the Torrance Tests of Creative Thinking?"
Psychology in the Schools, ii, 1974, 56-58.

 

Halpin, G. 8 Halpin, G. "The effect of motivation on
creative thinking abilities." Journal of Creative
Behavior, 1, 1973, 51-53.

 

Harris, M. 8 Evans, R. "Models and creativity." Psy h-
ological Reports, 3g, 1973, 763-769.

 

Harris, M. 8 Evans, R. "The effects of modeling and
instructions on creative reSponses." Journal of
Psychology, 82, 1974, 123-130.

 

 

Horner, M. "Toward an understanding of achievement-related
conflicts in women." Journal of Social Issues, Zfi,
1972, 157-175.

 

Horner, M. "Sex differences in achievement motivation and
performance in competitive and non-competitive
situations." Unpublished doctoral dissertation,
University of Michigan, 1968.

Hurlock, E.B. "The use of group rivalry as an incentive."
Journal of abnormal psychology, 22, 1927, 278-290.

Jersilid, A.J. Child psychology. 7th edition. Englewood
Cliffs, New Jersey: Prentice-Hall, 1975.

 

102

Johnson, D.W. 8 Johnson, R.T. "Instructional goal structure:
COOperation, competition, or individualistic."
Review of Educational Research, 13, 1974, 213-240.

Johnson, R. "Differential effects of reward versus no
reward instructions on the creative thinking of two
economic levels of elementary school children."
Journal of Educational Psychology, 99, 1974, 530-533.

Jones, S. 8 Vroom, V. "Division of labor and performance
under cooperative and competitive conditions."
Journal of abnormal 8 social psycholpgy, p1, 1964,
313-320.

Julian, J. 8 Perry, F. "COOperation contrasted with intra-
group and inter-group competition." Sociometry, £9,
1967, 79-90.

 

Kakkar, 8. "Effectiveness of motives." Manas: A Journal
of Scientific Psychology, lg, 1968, 59-66}

 

Kr0p, H., Alegre, C., 8 Williams, C. "Effect of induced
stress on convergent and divergent thinking."
Psychological Reports, 25, 1969, 895-898.

Kruglanski, A., Riter, A., Arazi, D., Agassi, R., Montegio,
J., Peri, I., and Peretz, M. "Effects of task-
intrinsic rewards upon extrinsic and intrinsic
motivation." Journal of Personality and Social
Psychology, El, 19 5, 699-705.

 

Leith, G. "The relationships between intelligence, per-
sonality, and creativity under two conditions of
stress." British Journal of Educational Psychology,
1;, 240-247.

Leonard, G. Education and Ecstasy. New York: Dell, 1968.

Levy, L.H. "Originality as role-defined behavior."
Journal of personality and social psychology, 3,
1968,872-78.

Maccoby, E. 8 Jacklin, C. Psychology of Sex Differences.
Stanford University—Press, 1974.

Maller, J. "Cooperation and competition: an experimental
study in motivation." Teachers College Columbia
University Contributions to Education,71929, NO. 384.

 

103

Manske, M. 8 Davis, G. "Effects of simple instructional
biases upon performance in the unusual uses test."
Journal of General Psychology, 12, 1968, 25-33.

 

May, M. 8 Doob, L. "Cooperation and competition." Social
Science Research Council Bulletin, 1937, No. 25.

 

Mendelsohn G.A. 8 Griswold, B.B. "Assessed creative
potential, vocabulary level, and sex as predictors of
the use of incidental cues in verbal problem solving."
Journal of personality and social psychology, 3, 1966
423-431.

 

Mischel, W. "Sex-typing and socialization." In P. Mussen
(Ed.), Carmichael's Manual of Child Psychology,
John Wiley 8 SOns, 1970, volume 2.

 

Neill, A.S. Summerhill. New York: Hart, 1960.

 

Parnes, S.J. 8 Meadow, A. "Effect of brainstorming
instructions on creative problem solving by
trained and untrained subjects." Journal of
Educational Psychology, £9, 1959, 171-176}

 

 

Peretti, P. "Effects of noncompetitive, competitive instruc-
tions, and sex on performance in a color-word inter-
ference task." Journal of Psychology, 32, 1971,
67-700

 

Piaget, J. "DeveIOpment and learning." In R. Rippley 8
V. Rockcastle (Eds.), Piaget rediscovered. Ithaca,
N.Y.: Cornell University Press, 1964, pp. 7-20.

 

Raina, M. "A study into the effect of competition on
creativity." Gifted Child Quarterly, lg, 1968, 217-
220.

Raven, B. 8 Eachus, H. "COOperation and competition in
means-interdependent triads." Journal of abnormal 8
social psychology, 91) 1964, 307-316.

 

 

Richman, M. "The effects of individual and group competi-
tion on productivity and interpersonal relationships
in elementary school children." Dissertation
Abstracts International, (33-2B), ,

 

Rogers, C. "Toward a theory of creativity." ETC: A review
of General semantics, ll, 1954, 249-260.

 

 

104

Shaw, M. "Some motivational factors in cooperation and
competition." Journal of Personality, 29, 1958,
155-169.

 

Sims, V. "The relative influence of two types of motivation
on improvement." Journal of Educational Psychology,
lg, 1928, 480-484.

 

Sommers, W. "The influence of selected teaching methods on
the deveIOpment of creative thinking." Unpublished
doctoral dissertation, University of Minnesota,
1961.

Stein, M. Stimulating Creativity. New York: Academic Press,
1974.

 

Stoneman, Z. 8 Keilman, P. "Competition and social stimula-
tion effects on simple motor performance of EMR
children." American Journal of Mental Deficiency,
IE: 1973, 98-100.

 

Torrance, E.P. "Creativity in the educational process."
In G. Lesser (Ed.), Psychology and educational
C

ractice. Scott, Fofesman, o., 1971, pp. 217-
341. .

 

 

Torrance, E.P. "Influence of dyadic interaction on creative
functioning." Psycholeglcal Reports, gs, 1970,
391-394.

Torrance, E.P. "Examples and rationales of test tasks for
assessing creative abilities." Journal of Creative
Behavior, 2, 1968, 165-178.

 

Torrance, E.P. Torrance Tests of Creative Thinking: Norms
TechnicalTManual. Princeton, New Jersey: Personnel
Press, 1966.

 

 

Torrance, E.P. Rewarding Creative Behavior. Englewood Cliffs,
New Jersey: Prentice-Hall, 1965.

 

Torrance, E.P. Explorations in creative thinking in the
early school years: II. An experiment in training
and motivation. iMinneapoliE? ’Bureau of educatifinal
research, UniVersity of Minnesota, 1959.

 

 

 

Torrance, E.P. 8 Aliotti, N. "Sex differences in levels of
performance and test-retest reliability on the
Torrance Tests of Creative Thinking Ability."
Journal of Creative Behavior, 3, 1969, 52-57.

 

105

Torrance, E.P. 8 Hansen, E. "The question-asking behavior
of highly creative and less creative basic business
trachers identified by a paper-and-pencil test."
Psychological Reports, l1, 1965, 815-818.

 

Tortorella, W. "The effects of a stressful situation on
a creative task." Dissertation Abstracts Interna-
tional, ge (3-B), 1965, 1214-1215.

 

Triplett, N. "The dynamogenic factors in pace-making and
competiting." American Journal of Psyphology, 9,
1898, 507-533.

 

Vernon, P.E. "Effects of administration and scoring on
divergent thinking tests." British Journal of
Educational Psycholpgy, gl, , -

 

Ward, W., Kogan, N., 8 Pankove, E. "Incentive effects in
children's creativity." Child Development, 1;,
1972, 669-676.

 

Whittemore, I. "The competitive consciousness."
Journal of Abnormal Psychology, 29, 1925, 17-33.

 

Whittemore, I. "The influence of competition on performance:
An experimental study." Journal of abnormal and
socialpsychology, lg, 1924, 236-253.

 

 

Zuckerman, M. and Wheeler, L. "To dispel fantasies about
the fantasy-based measure of fear of success."
Psychological Bulletin, 8g, 1975, 932-946.