4
I'lvu . I- ”I‘m. "
I-., .

-: -" Jiffi

s—

--- awe:

-._AP

1’

.- _l~o’o_-_ ... - - .‘A _
- $33 3'1gvo'af’: . :v’

.\'.' j.
. 1-, 9.

I' "$.14 .

J.

I

‘II. D‘ I O
.133. ,5";

\rxln ‘
1'1
‘ I

I
.I

I

'I I.
{I ,,
;,'_ IEII'.._'

’4“ _

.' . I‘IEITII
‘ ":‘Ifi‘

\I
>IK ' I
"11’. I»...

L","I II n .

V 0

.+

:K?
".‘.'.-.I.I I'I'I"'_

.I.‘,. i"
If I IO’IS'G‘\|IV“‘
WEI-1 ‘ K w;-

. I .1'
.I‘ '1 I'm”. ‘
P'“ ”3'1"

..,.....;,II
'5'

I I

I. .
(5' “H“fihh I‘f‘:1|\':"h
"thyme.” ,I-jI'II :Jf‘ ‘ {325.4
'I fII'jIIil‘IZ-gi.
.. III, I...

. AL.

“2%

I
.‘ZvJIuv' u II 2L1”

{3"}"I‘r-r95-
E

m

‘l‘."

:Ir‘4'1
l\‘

HM (TI'JL: " wri‘“
Y_. M1 33;! “.33
4‘ I.

K" '»‘I

. 1.2;}; . .

Iv: w: .
.- .C-.'.'- A
.I ,"1"'-
' ' {J

' £7.35
":l¢‘?”

r WWII; II, R 3%: .
€39...
31's.. "figfifl
. 9’43? v
“"3. “>352“.

‘1
(m.

I

.,,';‘I'.'d'o.- ha

I“
":I_\‘n
'.

.a

1:7") 3;
n i ‘

513' 7r...

‘ ' ' - .I

“3'. ”IN 3:" ."5-33?
I VI“ l...
“I“... t‘m.

$.33 $?-.J\"ALI‘V
:‘Y '9'“ "'1‘
I I I
. .
.

Utz‘.

§w
5.. am: ‘
H. :3.-

.' .V . I I"
{J "- ilk-"E“ {
«av r.

V5535“... .

.- 04'

"xi I.

”1
1
"71":

'1' I J

(III:
II

.I I“? {I
.‘IIO‘SI .3 ‘ .HI" ?::::|
‘ ‘j;rl’ .%?ra‘

I’lI‘f‘

"3.51:3

Via

"I. 5
I law‘IfiI-"V‘ III...“

gr? ‘II_:".'\ a"
‘ 434'. 2:" "

mfi? }:':'g%’y3§::¢‘ '33:}:34‘6
.QF~V'T‘..'.*.YF ‘14.“ '::E&
132,:‘2‘ {1 £ '.'.F‘ '(

l-cI-o
c

I'm.

V‘
II

LEI'S. u
If”:
\ ’3'"?

‘.' ‘r:
"I: ‘~‘

' .‘N'hn' I
9-054 :4? (ED
Slvi‘ w’l';
9“va 'I'(

 

WIS: 334:“.

IIIIIIIIIIZIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIII

10343 8044 L «...,
M v

'fl‘Q’:

Eben?»
Wm... . (mg,

This is to certify that the

thesis entitled

PLAY AND CREATIVITY: A FURTHER
EXAMINATION OF THE ASSOCIATIVE
FLUENCY HYPOTHESIS

presented by

Stephen A. Truhon

has been accepted towards fulfillment
of the requirements for

Ph. D. degree in Psychology

 

 

Date November 10, 1978

 

0-7639

OVERDUE FINES ARE 25¢ PER DAY
PER ITEM

   
   

Return to book drop to remove
this checkout from your record.

(

 

 

 

PLAY AND CREATIVITY: A FURTHER
EXAMINATION OF THE ASSOCIATIVE

FLUENCY HYPOTHESIS

BY

Stephen A. Truhon

A DISSERTATION

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

DOCTOR OF PHILOSOPHY

Department of Psychology

1978

0

ABSTRACT

PLAY AND CREATIVITY: A FURTHER
EXAMINATION OF THE ASSOCIATIVE
FLUENCY HYPOTHESIS

BY
Stephen A. Truhon

Recent research has found that children who are
given an opportunity to play display more creativity,
defined by the number of alternate uses for common
objects, than children who are not. The explanation for
this has been that play creates a mental set which per-
mits the player to associate fluently. This associative
fluency then enables the player to perform better on
creativity tests.

These studies have been confined to examining
verbal creativity. The question remains how play might
affect nonverbal creativity and personality characteris-
tics of creative persons, such as liking for complexity.
The first purpose of the current study was to examine
this question.

In addition, the associative fluency hypothesis
does not explain what aspect of play contributes to the

mental set. By a close examination of what children in

Stephen A. Truhon

the play condition do, the current study proposed to
answer this question and suggest a model to explain the
relationship between play and creativity.

Ninety kindergarten aged children (45 boys, 45
girls) were randomly assigned to either the play, imita-
tion, or control condition, in which they played with a
set of toys, imitated the experimenter in certain tasks
with toys, or colored pictures. They were then given a
battery of creativity tests--verbal, nonverbal and liking
for complexity--the order of which was systematically
varied,and a measure of intelligence. The children in the
play condition were also observed by observers who re-
corded at 15 second intervals the toys the child was
playing with, the integration of the play, and the
imaginativeness of the play, and rated the playfulness
of the child.

It was hypothesized that (1) children in the
play condition would perform better on the verbal
creativity test, worse on liking for complexity, and
the same on the nonverbal creativity test as; children in
the other two conditions, even when intelligence was
statistically controlled; (2) time spent playing with a
toy would be unrelated to the number of uses for that
toy; (3) complexity, integration, and imaginativeness in

play would be related to creativity; and (4) playfulness

Stephen A. Truhon

and creativity would be related, but not when intelli-
gence was partialled out.

In general, the hypotheses were not confirmed.
Girls in the play condition performed better on the verbal
creativity test than girls in the other two conditions,
but this was not true of the boys. There was a treat-
ment by order interaction that indicated that children
in the play condition performed best on the verbal
creativity test when it was given second, while those
in the other conditions performed best when it was given
third. Time spent playing with an object was unrelated
to the number or uses for that object. Complexity,
integration, and imaginativeness were near significance
when correlated with creativity. Playfulness and
creativity were related, even when intelligence was
partialled out.

A path analytic model that related playfulness
play and creativity was proposed. Playfulness-fun had a
strong positive relationship to shifts and complexity in
play, while playfulness-intelligence had a strong nega—
tive relationship with shifts in complexity. Verbal
and nonverbal creativity were weakly related to play.

The results of the current study indicate that
the relationship between play and creativity is more

complex than originally thought. Playfulness appears

Stephen A. Truhon

to be a highly useful construct in describing a child's
play and it probably has two aspects: fun and intelli-

gence.

ACKNOWLEDGEMENTS

I wish to epxress my appreciation to the following
persons who made this disseration possible:

Helen Benedict, chair of my committee, for allow-
ing me to sit in on her course, from which I developed
the topic for the current study, and for her help in
playing with ideas of the study;

John Paul McKinney for his invaluable help and
encouragement in my graduate training;

Elaine Donelson for her inspiring assistance in
the topic of creativity;

Richard Dobes for his statistical assistance and
for helping to reprogram my knowledge of FORTRAN;

Leighton Price for his useful comments in his
capacity as the College of Social Science's Dean's
Representative;

Bill Richards and Bob Wagner for being keen
observers of the phenomenon of play;

the kindergarten children, teachers, aides and
principles of the Cornell School, Okemos, the North
Aurelius School and the Steele Street School, both in
Mason, the Rusurrection School, Lansing, and the Memorial
School, Williamston for participating in the study;

ii

Sue Farhat, Sue Thill, and Cindy Mannon for their
help in the pilot study;

the kindergarten children, teachers, and principal
at the Dimondale School, Dimondale for participating in
the pilot study;

the Behavioral Science Instruction Laboratory
for the use of the programs SCRIBE and EDITOR which
enabled me to write earlier drafts of this dissertation;

Jack Condon, who seemed to be learning computer
operations just when I needed, for his help;

and Anthony and Beatrice Truhon, my parents,

for being the first people to play with me.

iii

TABLE OF CONTENTS

L I ST OF TABLE S O O O O O O O O O O 0

LIST OF FIGURES . . . . . . . . . . .

Chapter
I INTRODUCTION . . . . . . . . . .

The Cognitive Function of Play
Piaget's Theory . . . .
Sutton-Smith's Critique .
Singer's Approach . . .
Vygotsky's Theory . . .
Resolution of Theories on the C

Function of Play . . . .
Empirical Support for the Cognitive
Function of Play . . . . .

What is Play? . . . . . . . .

What Kind of Play is Related to
Creativity . . . . . . .

Creativity: Problems of Definition
and Measurement . . . . . .

Creativity: Problems with Young
Children . . . . . . . .

Play and Creativity: What Relation? .
Art and Play . . . . . . .
Creativity as Regression in the

Service of the Ego . . . .
Lieberman' 5 Theory of Playfulness .
Dansky and Silverman' s Associative

Fluency Hypothesis . . . . .
Other Studies of Creativity and Pla
Design of the Current Study . . .

Hypotheses . . . . . . . . .

o. . .. .
.“é.....

I I METHOD 0 O O O O O O O O O O O

subjeCts I O O O O O O O I 0
measures 0 O O O O O O O O 0
materials 0 O O O O O O O O 0

iv

Y

Page
vi

viii

t-uo G) \IU'le-‘H H

H P4 re
a. u»

NNH
0000

MN
00H

UDJNN
hNOO‘

Uh.)
\DkD

bu
N0

Chapter

Procedures . . . . .
Part 1 . . . . .
Play Condition . .
Imitation Condition
Control Condition .
Post-Tests . . .

Part 2 . . . . .

III RESULTS 0 O O O O O O O O O O O

The Effect of Play on Creativity:
Experimental Analysis . .' . . .
Measures Derived From Observations of
Play . . . . . . . . . . .
The Relationship Between Measures of
Creativity, Playfulness, and Play .
Multidimensional Scaling and Clustering
of Creativity, Playfulness, and Play
Path Analytic Model of Playfulness, Play,
and Creativity . . . . . . . .
Redundancy Analysis of Playfulness and
Creativity . . . . . . . . .

IV ‘DISCUSSION . . . . . - - ° ° ' '.

REFERENCE NOTES . . . . . . . . . . .
MFERENCES O O O O O O O O O C O O 0

APPENDIX A O O O O O O O O O O O O O

43
60
63
67
70
79
88

105
108
114

Table

10

11

LIST OF TABLES

Reliability of Judges on Play-—Objects
and Descriptions . . . . . . . .

Reliability of Playfulness Scale . . . .

Multivariate Analysis of Variance for the
Effects of Treatment on Measures of
creatiVity O O O O O O O O O O

Multivariate Analysis of Variance for the
Effects of Order on Measures of
Creativity . . . . . . . . . .

Multivariate Analysis of Variance for the
Effects of Sex on Measures of
Creativity . . . . . . . . . .

Multivariate Analysis of Variance for the
Effects of Sex by Order Interaction
of Creativity . . . . . . . . .

Multivariate Analysis of Variance for the
Effects of Treatment by Sex Interaction
on Measures of Creativity . . . . .

Multivariate Analysis of Variance for the
Effects on Treatment by Order Inter-
action on Measures of Creativity . . .

Multivariate Analysis of Variance for the
Effects of Treatment by Sex by Order
Interaction on Measures of
Creativity . . . . . . . . . .

Multivariate Analysis of Covariance for
the Effects of Treatment, Order, and
Sex of Creativity . . . . . . . .

Selected Correlations Between Measures of
Creativity, Playfulness, and Play . .

vi

Page

51
52

54

54

55

55

56

56

57

61

65

Table Page

12 Clusters Formed from the Creativity,
Playfulness, and Play Variables . . . . 72

13 Correlations Corrected for Attenuation
Between Clusters of Creativity,
Playfulness, and Play . . . . . . . 74

14 The Basic Assessment of the Path Analysis of
Playfulness, Play, and Creativity . . . 77

15 Correlations Among Measures of Playfulness
and Creativity . . . . . . . . . . 80

16 Factor Loadings of Playfulness and Creativity
on Redundancy Variates . . . . . . . 82

17 Correlation Matrices of Playfulness and
Creativity Redundancy Variates . . . . 83

18 Correlations Among Measures of Playfulness
and Creativity Mental Age Partialled . . 85

19 Factor Loadings of Playfulness and Creativity
with Mental Age Partialled on Redundancy
Variates . . . . . . . . . . . 86

20 Correlation Matrices of Playfulness and

Creativity Redundancy Variates with
Mental Age Partialled . . . . . . . 87

vii

LIST OF FIGURES

Figure Page
1 Recording Sheet . . . . . . . . . . . 46

2 Adapted Version of Playfulness Scale
(Lieberman, 1977) . . . . . . . . . . 48

3 Multidimensional Scaling Solution of Creativity:
Playfulness and Play Data . . . . . . . 68

4 Clustering of Creativity, Playfulness, and
Play Data . . . . . . . . . . . . 69

5 Multidimensional Scaling Solution and Clusters
for Creativity, Playfulness, and Play Data . 71

6 The Path Analytic Model for Playfulness, Play,
and Creativity . . . . . . . . . . . 7S

7 The Path Analytic Model for Playfulness, Play,
and Creativity with the Ordinary Least
Squares Estimates of Path Coefficients . . . 78

8 Mean Performance on the AUN given Time Spent
Playing . . . . . . . . . . . . . 94

9 Mean Performance on the AUU given Time Spent
Playing . . . . . . . . . . . . . 95

viii

CHAPTER I

INTRODUCTION

The purpose of the current study was to examine
the relationship between play and creativity. Since
creativity is considered by many researchers to be one
of a number of cognitive skills, it seems best to
examine first the relationship between cognition and
play. What the relationship is is not easily agreed
upon, as will be seen. By presenting the various
theories and noting where they overlap and contradict
each other it is hoped that some sort of resolution of
this relationship can be proposed that will lead to an
understanding of the relationship between play and a
specific cognitive skill, creativity. Since much of
the work on play and cognition originates with Piaget's

work, it seems appropriate to begin with him.

The Cognitive Function of Play
Piaget's Theory
According to Piaget (1962) there are two biologi-
cally based ways in which we adapt to the environment:
assimilation and accommodation. Assimilation, best

exemplified by play, occurs when the child adapts

elements in the environment to fit existing schema.
Accommodation, best exemplified by imitation,occurs
when the child adapts schema to fit elements in the
environment. Neither process exists alone, but each
always interacts with the other. Play, thus, plays an
important role in the development of the intellect by
encompassing new objects and events--a point with which
others agree (Klinger, 1969; Slobin, 1964).

Piaget classifies play into three types:
practice,symbol, and rule. Practice games begin in the
first months of life. They essentially involve sensori-
motor activities which lead to the acquisition of
skills. Symbolic games beginning about two years of
age imply the representation of an absent object. This
is the period of make-believe or pretend play. The
third type, rule games, rarely appears before seven
years. This type of play implies relationships with
others.

Both practice play and symbolic games disappear
with age, combining with other forms of play. Just as
persons become more logical and reality oriented as they
develop, so too their play becomes more reality based

and gradually disappears (also see Piaget, 1965).

Sutton-Smith's Critique

 

Sutton-Smith (1966) offered perhaps the first
major criticism of Piaget's theory of play, a criticism
which presented a number of points in rebuttal to his
theory. First, he felt that too great an emphasis is
placed on imitation in Piaget's theory of cognitive
develOpment so that play becomes subordinate to imitation
(a misinterpretation as Piaget [1966] points out).
Second, he argued that Piaget places too much emphasis
on play's relationship to reality based thought. "It
would not be far-fetched to speculate, in fact, that if
there is a relationship between play and thought, it is
more likely with . . . divergent intellectual Operations
[i. e., creative thinking] than with the directed
forms which concern Piaget" (Sutton-Smith, 1966,

p. 107).

Piaget would disagree with this viewpoint. In
discussing Groos's (1898) theory that certain play
activities prepare or pre-exercise adult skills, Piaget
(1962) concludes that symbolic games are not a pre-
exercise of imitation for later imaginative thought,
because imagination is not a cognitive skill, but a
cognitive process (i.e., imagination can be equated with

assimilation). Only indirectly can play lead to

imaginative thought. "Symbolic play will only achieve
its final form of creative imagination provided it is
as it were reintegrated in thought as a whole" (Piaget,
1962, p. 155).

Piaget suggests that the relationship between
play and imagination is a very weak one. He maintains
that symbolic games contribute to the cognitive process
as a whole (ultimately as reality based thought),
because, due to the child's egocentrism, the symbol
expresses present reality.

Sutton-Smith's (1966) third criticism concerns
Piaget's notion that play all but disappears in late
childhood. By doing this Piaget ignores the influence
of play on later development. In Sutton-Smith's view,
play does not disappear but differentiates into other
functions such as determining outcomes in counting out
games, ritualizing dramas of success and failure in
sports, and socializing adults into society's norms.
However, to Piaget (1966) this differentiation is the
same as disappearance, since play becomes more adaptive
to reality.

Sutton-Smith (1967) has elsewhere elaborated
his thoughts about play and cognitive development.
Play is involved in such activities as consolidating
learnings, preventing their disuse, producing creative

ideas, categorizing and learning skills such as number

conservation. This involvement is tenuous because of
the nature of play. Play would not be play if it were
strictly utilitarian.

Sutton-Smith (1971),in replying to Piaget's
(1966) response, marks a crucial difference between
Piaget and himself. Because of his cognitive orienta-
tion Piaget tends to look for aspects in play which
correspond to his theory. As a result, Piaget's approach
to play has to be subordinate to that of cognitive
development. Sutton-Smith, on the other hand, starts
from play and moves from it in examining cognition.
This difference in orientation plus Sutton—Smith's
emphasis on the role of play in adolescent development
are the major distinctions between Piaget's and Sutton-

Smith's approaches to play and cognition.

Singer's Approach

Much of Singer's (1966, 1973) work parallels
points made by Sutton-Smith. His examination of such
topics as fantasy and daydreaming suggests that play has
an important role in the years after childhood. In his
view, play gradually becomes internalized, because of
greater cognitive capacity, including memory, and
pressures against fantasizing out loud. One common form

of internalized play is daydreaming.

According to Singer (1966) daydreams can be
categorized along three dimensions: simple-complex,
reminiscent-planful, and personal-impersonal. The
simple-complex dimensions refers to the intellectual
intricacy of the daydream. The personal-impersonal
dimension measures the degree to which the daydream
involves the daydreamer. Most important for the purposes
of the current study is the reminiscent-planful dimen-
sion which involves the degree to which the daydream is
past- or future-oriented.

Instead of viewing daydreaming as wasteful, as
it commonly is, Singer suggests that daydreams, when
they are planful, can serve an important cognitive
function, problem solving. Through daydreaming one
attempts to pose solutions to matters that concern a
person, whether in planning a dinner party or deciding
what to do on a weekend.

One would expect that when problems increase so
might daydreaming. A rise in daydreaming during
adolescence could then be explained by the changes and
the problems of this period (Singer & Antrobus, 1963).

It appears that there is a general agreement
between Sutton-Smith and Singer that play has a role in
cognitive development after childhood. The only differ-
ence is that Singer has emphasized one aspect of this

play, daydreaming.

Vygotsky's Theory

 

Vygotsky (1967), outside the influence of
Piaget's work on play, provides some insights that may
help resolve the questions raised by Piaget, Sutton-
Smith, and Singer. According to Vygotsky there are two
important aspects of play: all play deals with imaginary
situations and all play has rules. For vygotsky each
of these aspects implies the other. By this Vygotsky
means that all play is somewhat removed from reality and
within this play there are certain behaviors that are
permitted or not permitted, even if they have not been
formally prescribed.

In contrast to Piaget (1962), Vygotsky feels
that play becomes less reality based during the period
of childhood. This is important for cognitive develop-
ment because the child learns not to rely solely on
external objects in the environment. Meaning is no
longer determined by an action (e.g., stomping one's
foot does not have to signify anger, but can signify
starting an imaginary motorcycle) or by an object (e.g.,
a stick can be a horse).

According to Vygotsky, play permits cognitive
development in a manner similar to Piaget's notion of
assimilation. Vygotsky agrees with Piaget (1965) that

through play the child begins to distinguish physical

rules from moral rules, particularly rules of self-
restraint. Finally, through use of imaginary situations
the basis for abstract thought may be developed.

Resolution of Theories on the
Cognitive Function of Play

 

 

From these theories the question is raised: how
is play related to imaginative thought in the individual%s
life—span? Piaget proposes a weak relationship based
upon the observation that there is less play as one
grows older and that which occurs is more reality based.

The key word here is "observation." Less play
is visually apparent as a child grows older, but what
is occurring in the child's mind? A child's thoughts
are less constrained by reality as the child grows
older, a point with which even Piaget would agree (e.g.,
the ability to consider the possible rather than the
actual in formal operations).

It does not seem likely that a set of behaviors,
such as play, would just disappear, as Piaget suggests.
A transformation seems more likely. Daydreaming, as
Singer's work strongly suggests, may be one form of
transformed play.

It is generally agreed upon that a relationship
between play and cognition, in particular--imagination,

exists. Play can aid imagination, but under what

conditions is unclear. Perhaps this can be resolved
by examining experimental studies of play and cognitive
functioning.

—Empirica1 Support for the
Cognitive Function of

13121

 

The relationship between play and cognitive
functioning has been examined in an indirect manner
among socially disadvantaged children. Smilansky (1968)
noted that children from underprivileged homes did not
play properly: their play was either monotonously
repetitious or disjointed, jumping from topic to topic
without organization. Since these are the children
who also have scholastic problems later on, Smilansky
puzzled about the connection between these two kinds of
problems.

To one group of these disadvantaged children
Smilansky began teaching techniques of sociodramatic
play, i.e., pretend games which involve real life
situations. After nine weeks of treatment she found
that these techniques brought about greater socio-
dramatic play as well as greater persistence, social
interaction, and communication.

It can be argued that the changes brought about
by the sociodramatic training in Smilansky's study are

not cognitive but social. However use of social and

10

cognitive in such dichotomized terms obscures the point
that skills such as persistence, interaction, and
communication have both social and cognitive implica-
tions. In addition Smilansky's study has influenced
other researchers who have examined the effect of socio-
dramatic training on more strictly cognitive skills.

Several other studies (Freyberg, 1973; Rosen,
1974; Saltz & Johnson, 1974) have confirmed the lower
levels of sociodramatic play among children from lower
SES backgrounds. Training these children in the tech-
niques of dramatic play has led to greater imaginative-
ness, positive affect, concentration (Freyberg, 1973),
better performance on group cooperative tasks (Rosen,
1974), and improvement in intelligence and interpersonal
perception (Saltz & Johnson, 1974).

A somewhat different approach to the cognitive
function of play is demonstrated in a study by Sylva,
Bruner, and Genova (1976). In their study children were
required to remove a piece of chalk from a box which
was more than an arm's length away. The solution was to
clamp several sticks together, which could pull the box
closer and remove the chalk.

Prior to this problem different training condi-
tions were used: an observation condition, in which an
adult constructed an elongated tool with clamps and

sticks; a play condition, in which the children played

11

with clamps and sticks; an observe components condition,
in which the children saw a puppet show with illus-
trated the techniques necessary for solution of the
problem; a training of components condition, in which
while manipulating the clamps and sticks the children
were given specific instructions; and a no treatment
condition.

Children in the play and observation conditions
were more likely to solve the problem spontaneously than
the other groups. Further analysis revealed differ-
ences between children in these two conditions. Children
who played made more effective use of hints, varied
their approach more, and this approach was varied more
systematically. Sylva et al. attributed these differ-
ences to the fact that the children who played had to
initiate the solution themselves.

From these studies it is clear that play is
related to such cognitive skills as imagination and
problem-solving. To clarify the relationship between
play and creativity, however, two additional questions
must be answered: (1) What is play? and (2) What kind

of play is related to creativity?

What is Play?

 

The definition of play has frustrated many in-

vestigators of play. The wide diversity of activities

12

called play make such a definition seem so impossible
that Garvey (1977) rephrased the question as "What are
play?"

One way of avoiding the problems of defining play
is to list commonly accepted features of play. These
features include: (1) its purposelessness, i.e., it
is an end, in and of itself; (2) the pleasure it brings;
(3) that it is usually exhibited by immature organisms;
and (4) that it involves some distance from reality.
While these features appear useful, one can think of
activities that would violate any one of these features
and still be called play. (As a result, one should
view these features as not very strict guidelines.)

With these features in mind, play will be de-
fined in the current study as behaviors which are
intrinsically motivated, engaged in for "their own sake,"
dominated by the organism--not outside forces, and
accompanied by positive affect. This definition was
chosen because it was consistent with the features of
play. In addition, this definition conceptualizes play
as non-utilitarian but allows play, at the same time,
to have functional application to other skills, such as
cognition. Thus play can be related to creativity but
does not have to be. As a result, the current study
can examine what aspects of play may be related to

creativity.

13
What Kind of Play is Related
to Creativity?

In contrast to the difficulty of defining play,
categorizing types of play appears relatively easly.
There is general agreement on three categories of play:
practice play, symbolic games, and rule games (Piaget,
1962).

Before we can relate one of these categories of
play to creativity it is important to compare creativity
with other cognitive skills. Of those skills discussed
thus far, creativity seems most similar to imagination.
Certainly imagination is an important aspect of the
creative process. From this it is reasonable to assume
that the relationship between play and creativity should
parallel that between play and imagination.

Of those types of play mentioned above, it should
be apparent that symbolic games are closely related to
imagination, especially in the studies on the effects of
training children from lower SES backgrounds in tech-
niques of sociodramatic play. If then creativity is
related to symbolic games, it should be relatively easy
to look at the amount of time spent in symbolic games
and the type of symbolic games played and correlate
them with creativity scores.

But problems immediately occur. What is creativ-

ity and how is it measured? Unless these questions are

l4

effectively answered, conceptualizing a relationship

between symbolic games and creativity will be difficult.

Creativity: Problems of Definition
and Measurement

 

Since Guilford (1956) and his associates have
been responsible for much of the recent research in the
area of creativity, it is appr0priate to begin a con-
sideration of the definition and measurement of creativity
with his work. According to Guilford's Structure of
the Intellect Model there are three important attributes
to any intellectual activity: the cognitive process,
the context to which the process is applied, and the
product of this process. Two kinds of cognitive pro-
cesses are of interest here: convergent thinking, which
involves focusing upon one correct answer, and divergent
thinking, which involves searching for a number of
answers in a broad based manner. Divergent thinking
corresponds closely to Guilford's (1957) idea of
creativity.

Much of the work of Guilford and his associates
has been in developing tests to measure the attributes
of intellectual activity, including divergent thinking.
Three general divergent thinking abilities have been
noted by Guilford: fluency, the ability to generate
many ideas in a short time; flexibility, the ability to

vary these over a wide range, i.e., not in one category;

15

and originality, the ability to generate unique or un—
usual ideas (see Wallach's [1970] review).

Torrance (1966a, 1966b) has adapted Guilford's
concepts for testing with children. Two types of tests
were develOped--verbal and figural. In addition to the
three general abilities noted by Guilford, Torrance has
added afourth-elaboration, the ability to add decor-
ative, unnecessary aspects to the ideas generated.

Some problems developed with the use of crea-
tivity tests such as those devised by Guilford and
Torrance. To argue for the existence of a skill one
must show that it is distinct from other skills. In
creativity research the distinction between creativity
and intelligence has been an important one.

Getzels and Jackson (1962) sought to show such
a distinction by demonstrating that differences existed
between those high in creativity, low in intelligence and
those low in creativity, high in intelligence. Reviewers
of this study noted that here, as well as with many
studies using Guilford's or Torrance's tests: (1) there
was no common characteristic to creativity tests, inter-
correlations were quite low (Thorndike, 1963); and (2)
creativity tests correlated with intelligence tests as
well as they did with other creativity tests (Burt,

1962).

16

An alternative view toward creativity has been
develOping. Mednick (1962) hypothesized that the
creative thinker had a flat gradient rather than a steep
one with respect to the likelihood of thinking of cer-
tain associates when given a task. This meant that the
creative thinker is more likely to come up with remote
or unusual associations of words or events. Thus,
according to Mednick's theory, creativity is defined as
an ability to associate previously unrelated ideas to
produce a IreW'and interesting object.

Building upon Mednick's theory Wallach and Kogan
(1965) developed their own tests of creativity. Although
similar to some of (3uilf0rd's tests, there were two
important differences: (1) to aid the association
process, no time limit was imposed; and (2) to contri-
bute to the playful aspect of creativity, these tests
were administered in a gamelike atmosphere on a one-to-
one basis. Two scores were calculated for each tests:
number (fluency) and uniqueness (originality). Corre-
lations among creativity measures were quite good and
the correlations between creativity and intelligence
measures were quite low.

Despite the apparent success of Wallach and
Kogan's research, some questions remained. Guilford
(1971) argued that unlimited testing time allowed the

children to develop strategies. Tests were tests and

17

should be administered as such. More valid is a point
raised by Anastasi and Schaefer (1971): creativity and
intelligence are multifaceted abilities and should over-
lap to some extent (not the near zero correlations of
Wallach and Kogan's [1965] research).

These criticisms raise important points in
defining creativity. A definition of a multifaceted
skill like creativity should not be too restrictive,
otherwise too much is lost. Defining creativity as an
associative process, in which previously distinct ele-
ments are combined to produce a novel result, is useful,
because such a definition can include the divergent
thinking abilities noted by Guilford and Torrance. How-
ever, such a definition needs to include a number of
personality characteristics, such as interest in change
and.a liking for complexity, which result from this
associative process.

Therefdre, for the purposes of the current
study, creativity was defined as the association of
elements, verbal and nonverbal, which were previously
distinct, in a novel way which results in a novel pro-
duct which fits in with and may also extend the bounds
of reality. In turn this associative process over a
period of time will manifest itself in a number of

personality traits, such as playfulness, openness, and

18

a liking for complexity, which will further aid the
associative process.

Such a definition was chosen for a number of
reasons. Defining creativity as an associative process
keeps it consistent with Mednick (1962) and Wallach and
Kogan's (1965) work, and thus distinct from intelligence.
Use of the words "verbal" and "nonverbal" emphasizes
the point that creativity involves activity that is not
always verbal. The second part of the definition per-
mits a developmental perspective to creativity. Second,
it notes that there are cognitive and personality
aspects to creativity and that they are interrelated.
This definition allows for a similarity between play and
creativity both in the way objects are combined and in
the playful attitude that results from creativity. At
the same time this definition does not force play and
creativity to be related just by definition.

Since the current study is concerned with play
and creativity, the population under consideration is
children. Are there specific problems in examining
creativity in young children?

Creativity: Problems with
Young Children

 

Starkweather (1964) has noted that further
problems crOp up when trying to measure creativity with

young children. With adults one can attempt to measure

l9

creative achievement, but with children one is forced
to measure creative potential. Still these tests base
performance on adult criteria. Also young children
respond differently to these tasks than older children,
e.g., they are less verbal. Finally, the different
aspects of creativity (i.e., fluency, flexibility,
originality and elaboration) may not be differentiated
in young children.

Based upon these potential problems, Starkweather
(1964, 1971) has developed her own measures of creativity.
Her tests are questionable validity, since they have no
tradition to fall back on. Since Starkweather denies
that fluency, flexibility, originality,and elaboration
are differentiated in early childhood, their validity
cannot be seen by comparing them with Guilford's tests.
Similarly there is no discussion of an associative pro-
cess that would enable a comparison with Wallach and
Kogan's. It would take a longitudinal study to confirm
her tests' validity.

Despite the problems forecast by Starkweather,
most studies of creativity with young children have
employed Wallach and Kogan's (1965) tests of creativity,
adapting some of the items for easier understanding by
children (e.g., Singer & Rummo, 1973; Singer & Whiton,
1971; Ward, 1968, 1969). While these tests appear

appropriate because they show an independence of

20

creativity from intelligence, they are not sufficient to
consider all aspects of creativity.

All of Wallach and Kogan's (1965) tests require
verbal responses. Torrance's (1966a, 1966b) work
demonstrates a weak relationship between verbal and non-
verbal creativity. In keeping with this finding and
the definition of creativity as an associative process,
involving verbal and nonverbal elements, studies of
creativity among young children should be more broad-
based.

Since a working definition of creativity has been
arrived upon and the problems of measurement with kinder-
garteners resolved, the relationship between play and

creativity can be examined in a systematic manner.

Play and Creativity: What Relation?

Art and Play

 

The idea that play and creativity are related is
certainly not a new idea. This can be seen clearly in
the early psychological writings about play, especially
those dealing with the relationship between art and
play. Groos (1898) noted that "all forces efficacious
in artistic production are referable to the central

idea of play" (p. 328). Similarly for Spencer (1899),

21

"the activities we call play are united with aesthetic
activities" (p. 627).

Psychoanalytic theorists were concerned with the
roots of art and saw it in play. "Perhaps we might say
that every child at play behaves like an imaginative
writer" (Freud, 1908-1959, pp. 173-174). Both art and
play were seen as attempts to relate the id to reality.
The theorist who best elaborated this relation was
Ernst Kris.

Creativity as Regression in the
Service of the Ego

Ac:cording to Kris (1952) there are three stages
in the development from play to art. In the first
stage the child begins to distinguish between reality and
make-believe. As a result, instead of doing what he/she
wishes impulsively, the child can play at or pretend to
do it.

The distinction between reality and make-believe
becomes more durable. In the second stage the child can
entertain two notions simultaneoulsy: a firm belief in
the "reality of play" coexisting with the certainty that
it is "only" play. In the third stage, the child can
accept the fantasies of others as well as his/her own.

The other aspect to Kris's theory is that
creativity is seen as an act of regression in the service

of the ego. In other words, regression, the return to

22

earlier ways of existing, usually a defense mechanism.is
controlled by the ego for a useful purpose. Through the
preconscious there is a shuttling between id material
and ego control, akin to the notions of inspiration and
elaboration in creativity.

This theory has come under attack for not being
specific enough. In Bellak's (1958) view creativity
involves the relative reduction of certain adaptive
functions in the service of synthetic ego functions.

In Piagetian terms, this means less accommodation and
more assimilation. For Bellak there is regression, but
not total regression.

Weissman (1967) finds the use of the term
"regression" inappropriate. Regression is a destructive
process and using it with creativity perpetuates the
myth that genius is related to insanity. He prefers
the term "dissociative functions." In this way the
ego can disrupt itself from established object relations
and can consider new ones (note the similarity to
vygotsky's [1967] point about meaning becoming separ—
ated from the object during play). Using dissociative
functions eliminates the overemphasis that Weissman
feels Kris placed on inspiration in the creative process.

One can see that there are a number of problems
with this theory. The problems increase when one tries

to examine the relationships between play and

23

creativity, because it is never explicitly stated in
Kris's theory. By inference it appears that during
the creative process a person regresses so that he/she
can combine or "play with" ideas, as one plays as a
child. Play is then a necessary condition for creativity,
but how much play or what kind of play will produce a
creative person is not considered. In this way it be-
comes rather difficult to test this hypothesis empiri-
cally. This difficulty is further demonstrated in the
lack of a relationship found between regression (as
measured by the Rorschach test) and creativity (as
defined by traditional measures such as the Torrance

tests) (Lazar, Note 1).

Lieberman's Theory of Playfulness

A somewhat different approach to the relation-
ship between play and creativity has been offered by
Lieberman. Lieberman (1977) has noted that in several
studies (e.g., Getzels & Jackson, 1962; Torrance, 1961)
playfulness has been a trait differentiating highly
creative persons from those less creative.

According to Lieberman, the child who enjoys
play and is afforded the opportunities to play, under
certain circumstances, will become playful. With the
internalization of play in later childhood, described

by Singer (1973), this playfulness survives and becomes

24

a personality trait. This playfulness enables
one to combine disparate thoughts to produce something
creative.

There are five elements to Lieberman's (1977)
concept of playfulness: physical, social and cognitive
spontaneity, manifest joy, and sense of humor. Based on
these elements Lieberman has developed scales for kinder-
gartners (Lieberman, 1965) and adolescents (Lieberman,
1967).

To support her theory, Lieberman (1965) rated 93
kindergarten children on her Playfulness Scale and tested
them on three measures of creativity. There were signif-
icant correlations between the elements of playfulness
and the fluency, flexibility, and originality scores on
the creativity tests. A study of Durrett and Huffman
(1968) replicated most of these findings, although there
was no correlation between playfulness and originality.

One problem with Lieberman's scale is its failure
to deal adequately with the two-dimensional character of
playfulness. Lieberman alludes to but never directly
confronts the issue that playfulness can be examined for
its applied value to cognitive skills and for its own
intrinsic or "fun" value. This duality is also seen in
the elements of playfulness. For example, in sense of

humor there is a cognitive component in order to

25

appreciate the joke as well as an affective component
for the laughter and smiling that accompany it.

This dual nature has been noted in other aspects
of play. Sutton-Smith (1967) has pointed out that play
aids cognitive development in a number of ways. But
if one looks only for what can be used from play, one
misses the pleasure that play brings. Similarly in
discussing the types of daydreams Singer (1966) dis-
tinguishes between future-oriented daydreams which are
usually aimed at problem solving and past-oriented day-
dreams which are fanciful.

A second problem is that of the relationship
of creativity and intelligence which resurfaces in these
studies, especially Lieberman's (1965). When factor
analyzing her data, Lieberman (1965) found that all the
variables in her study, including intelligence, loaded
heavily on one factor. This, as Sutton-Smith (1967)
pointed out, poses the question whether creativity and
playfulness are just two manifestations of intelligence.
In fact, Singer and Rummo (1973) discovered that when
mental age was partialled out, there was no relation
between creativity and playfulness in Lieberman's (1965)
study.

There are further questions about Lieberman's
work. Although her theory about playfulness seems

plausible, her methodology is open to question.

26

Lieberman (1977) demonstrated the validity of the Play-
fulness Scales by the fact that the five elements hang
together, apart from such "dummy" variables as achieve-
ment orientation and attractiveness when factor analyzed.
This type of analysis is not sufficient to establish
validity.

Campbell and Fiske (1959) have suggested that the
way to test the validity of a trait is by a multitrait-
multimethod analysis. In the multitrait aspect of the
analysis the researcher compares the hypothesized
measure of the trait with established measures (usually
at least one similar to and one theoretically unrelated
to the hypothesized trait) to determine its independence.
In the multimethod aspect of the analysis the researcher
examines various ways to measure the trait to make cer-
tain that its existence is not an artifact of the
method used. Certainly such an approach would be use-
ful in validating the Playfulness Scales.

Dansky_and Silverman's Associative
Fluency Hypothesis

 

A different theory about the relationship
between play and creativity which stresses the associa-
tive process in creativity has been offered by Dansky
and Silverman (1975). According to this hypothesis

play creates a mental set which enables the player to

27

generate associations to a variety of objects. Dansky
and Silverman point to three studies to support their
view.

Sutton-Smith (1967) had noted that children when
asked what they could do with certain toys, arrived at
more uses and.more unusual uses for toys whiCh they
preferred. However, there was a confounding here, as
(noted by Dansky and Silverman (1973), between playfulness
and exposure to the toys.

To eliminate this problem Dansky and Silverman
(1973) presented a set of toys to two groups. One
group was allowed to play with the toys, while the other
group had to imitate the activity of the experimenter
with these toys. A third group was given an irrelevant
(drawing) activity. The children in these three groups
were later asked to tell the experimenter all the
different ways in which one of the toys could be used.
The group who were permitted to play produced more stan-
dard and unusual uses for each of the toys than the
other two groups. In a similar study Dansky and Silver—
man (1975) found that children who had played also
produced more standard and unusual uses for toys they
had not played with than the imitation and control
groups.

There are certain similarities between this

hypothesis and those mentioned above. This ability to

28

associate fluently is similar to Vygotsky's (1967)
thoughts about the liberation of meaning from an object
during play. The difference between Dansky and Silver-
man's (1975) and Lieberman‘s (1977) may only be time:
If a mental set for associative fluency becomes long-
lasting, it may become a personality trait called play-
fulness.

A possible link between these theories can be
seen in a study by Feitelson and Ross (1973) that was
done over a seven week period. Kindergarten children
were assessed on several measures of creativity, including
the Cincinnati Autonomy Test Battery and the Torrance
Figural Tests. Two weeks after this assessment each
child was randomly assigned to one of four treatments:
play and tutoring; play, no tutoring; tutoring, no play;
and no play or tutoring. Half-hour sessions of the
treatment conditions were held twice a week for five
weeks. Children were then reassessed on the creativity
measures. Children who had been assigned to the play
and tutoring treatment showed the greatest gains in
creativity.

Two major questions are not answered by Dansky
and Silverman's (1973, 1975) studies. Their definition
of creativity is more restricted than the one developed
in the current study. Creativity is defined by-them as

a verbal associative process. It can then be asked

29

whether play aids creativity in the broad sense as
defined in the current study or only in the restricted
sense as employed by Dansky and Silverman.

The second question involves the mental set that
Dansky and Silverman (1975) hypothesized is created
during play. It remains unclear from their discussion
how play does this. By a closer examination of play this
problem may be clarified. But at this point it is
unknown what aspects of play contribute to this mental
set. Among the aspects of play already discussed

imaginativeness and playfulness are possibilities.

Other Studies of Creativity
and Play

Hutt and Bhavani (1972) examined the relationship
between style of play and creativity. They classified
three- to five-year-old children into three groups:
non-explorers, who looked at and approached a new toy
but did not inspect it; explorers, who actively investi-
gated the toy but did little else with it; and inventive
explorers, who, after investigating the toy, used it in
many imaginative ways. They then tested these children
with the Wallach and Kogan (1965) creativity tests.

They found that the inventive explorers scored higher
on the uniqueness scale than the other groups.

J. E. Johnson (1976) had judges categorize the

play of three- to five-year olds as either fantasy or

30

non-fantasy play. In addition, the fantasy play could

be described as either social or non-social. He also
administered intelligence tests and Ward's (1968) adapta-
tion of the Wallach and Kogan creativity tests. There
were significant correlations between the measures of
intelligence and creativity and social fantasy, although
greater for the measures of creativity. Neither intelli-
gence nor creativity was related to non-social fantasy
play.

BishOp and Chace (1971) examined the environ-
mental conditions that promote creativity in children.
Mothers were given a battery of personality inventories,
while the creativity of their children was assessed by
their play with geometric shapes. They found that
children of more abstract mothers, who also provided a
more playful environment, showed greater creativity in
the complexity of the shapes chosen, the variety of
colors used, and in the complexity in the final design
created. The implication is that mothers who are more
abstract are more likely to provide a playful environ-
ment which results in creativity in their children. But
it can also be argued that creative children are more
demanding about play, causing their mothers to become
more abstract and to provide a more playful environment.

One other way of looking at the relationship

between play and creativity has been through the study

31

of imaginary companions. In a series of studies of
creativity among adolescents Anastasi and Schaefer
(1969; Schaefer, 1969; Schaefer & Anastasi, 1968)

have noted that creative high school students were more
likely to have had imaginary companions as children
than their less creative counterparts. However,
Manosevitz, Fling, and Prentice (1977) found that young
children who have imaginary companions are not more
creative than those who do not. This difference in
results may be due to the internalization of play that
is necessary for creativity (cf. Singer, 1973).

In reviewing these studies two points should be
noted. The first is that while these studies describe
certain characteristics of play that are related to
creativity, they do not answer the questions whether
play aids creativity, when it is defined as other than
a verbal associative process, and what aspects of play
creates the mental set which leads to creativity.

The second point is conspicuous by its absence.
No sex differences have been reported in any of the
above mentioned studies. This is not surprising since
these studies have tended to avoid tasks and toys that
are stereotypically masculine or feminine. It is also
noteworthy that sex differences in creativity are not
usually found among children of ages testedixrthese

studies. Maccoby and Jacklin (1974) have reported that

32

on verbal creativity tests girls begin to show superior
performance about seven years of age, while on nonverbal
creativity no clear pattern of sex differences has been

found.

Design of the Current Study

 

It is clear that other studies of play and
creativity do not answer the questions raised by Dansky
and Silerman's (1973, 1975) studies. To answer the
first question--whether play aids only verbal associa-
tive fluency, or other aspects of the creative process
as well--required an experimental multivariate design.
Given the time constraints of working with kindergarten
aged children it was necessary that the creativity tests
be few in number yet varied.

To meet this specification a pilot study was
undertaken to determine the dimensionality of creativity
for kindergarten aged children. Three creativity tests,
the Alternate Uses Test (Wallach & Kogan, 1965), the
Barron-Welsh Art Scale (Barron & Welsh, 1963), and the
Torrance Figural Test (Torrance, 1966b), from which 13
scores were calculated, were administered to 60 kinder-
garten aged children. The scores were correlated and
submitted to multidimensional scaling (see Kruskal,

1964).

33

From this scaling two dimensions emerged. The
first dimension showed the Alternate Uses Test at one
end of the dimension and the Barron-Welsh Art Scale at
the opposite end of the dimension. The Torrance Figural
Test was represented on an independent dimension. The
flexibility scale from this test showed this most clearly.
These three tests were chosen for examination in the
experimental analysis.

In the current study Dansky and Silverman's
(1973, 1975) examination of the relationship between
play and creativity was applied to a broadbased view of
creativity. Thus, children were assigned to the play,
imitation, and control conditions for 10 minutes. But
instead of being given one creativity test, the three
mentioned creativity tests were administered, systemati-
cally controlling for order. In addition, a measure of
intelligence was also administered in order to control
for the effect of intelligence on creativity.

The second question remaining from Dansky and
Silverman's (1973, 1975) studies--what aspects of play
produce the mental set that results in creativity--
required an exploratory, correlational analysis. It
was heped that this question would be answered by
examining the children in the play condition and record-
ing at 15 second intervals what these children did

during play. In addition by rating these children on

34

Lieberman's (1977) Playfulness Scale further evidence
was thought could be gained about its validity. Then
measures derived from the observations of play and the
Playfulness Scale could be used to propose a model for
the relationship between play and creativity. Multi-
dimensional scaling, although not the only way to build
a model, seemed a logical first choice. Such a multi-
dimensional scaling model could then be compared with
one derived from Lieberman's (1964) data to determine

the generalizability of the model.

Hypotheses

 

Using this design certain hypotheses could be
formulated about the relationship between play and
creativity. Dansky and Silverman (1973, 1975) have
demonstrated that play aids verbal associative fluency.
But it was unclear how play affects nonverbal creativity
(as measured by the Torrance Figural Test) and liking
for complexity (as measured by the Barron-welsh Art
Scale), since studies of play and creativity have not
examined these aspects. However, there was information
from the pilot study that liking for complexity appeared
opposite of verbal associative fluency and that non-
verbal flexibility was independent of verbal associative
fluency. If these findings are applied strictly to

Dansky and Silverman's (1975) associative fluency

35

hypothesis, it was expected that in response the first
question raised from Dansky and Silverman's (1973, 1975)
studies, children in the play condition would perform
-better on a measure of verbal associative fluency, worse
on liking for complexity, and the same on nonverbal
flexibility as children in the imitation and control

conditions.

Hypothesis 1

 

Children in the play condition should perform
better on the Alternate Uses Test, worse on the
Barron-Welsh Art Scale, and the same on the
Flexibility Scale of the Torrance Figural Test
as children in the imitation and control
conditions.

Critics of studies on creativity have pointed
out that many of these studies (e.g., Getzels & Jackson,
1962; Lieberman, 1965) have been confounded by the
fact creativity is related to intelligence. It was
therefore advisable to control statistically for intelli-
gence in examining Hypothesis 1. But there is nothing
in Dansky and Silverman's (1975) associative fluency
hypothesis that assumes that intelligence influences the
process involved. It was then expected that Hypothesis

1 would not change, when intelligence was statistically

controlled.

Hypothesis 2

 

Children in the play condition should perform
better on the Alternate Uses Test, worse on

36

the Barron-Welsh Art Scale, and the same on the

Flexibility Scale of the Torrance Figural Test

as children in the imitation and control condi—

tions, even when intelligence is statistically
controlled.

The next group of hypotheses involve the second
question raised by Dansky and Silerman's (1973, 1975)
studies-~what aspects of play are related to what kinds
of creativity. This was the exploratory part of the
study and thus there was some uncertainty what the
model, which related aspects of play with kinds of
creativity, would look like. Nevertheless it was possi-
ble to hypothesize about the relationships between some
of the variables.

Dansky and Silverman (1975), in discussing the
finding that children in the play condition generated
more alternate uses than children in the imitation and
control conditions even for objects they had not played
with, suggested that it is play itself that creates a
mental set that enables one to generate uses for an
object, rather than contact with that object. That
suggested that the time spent playing with an object

would be unrelated to the uses for that object.

Hypotheis 3

 

Time spent playing with an object should be
unrelated to the number and uniqueness of
uses for that object.

37

A number of studies (e.g., Hutt & Bhavani, 1972;
Sylva et al., 1976) have noted that how a child plays is
related to the way the child creatively solves problems.
It was, therefore, hypothesized that measures derived
from the observation of play, which involved style of
play, (e.g., complexity, integration, and imaginativeness)

would be related to performance on creativity tests.

Hypothesis 4

 

Complexity, integration, and imaginativeness
of play should be related to creativity.

Finally, Lieberman's (1977) concept of playful-
ness required further examination. According to
Lieberman playfulness through the internalization of
play can be reflected in creativity. To support this
view there have been studies that have found a positive
relationship between playfulness and creativity (Durrett
& Huffman, 1968; Lieberman, 1965). But in demonstrating
the validity of her concept by factor analysis, mental
age was found to load highly on the same factor as the
items from her Playfulness Scale. This prompted
Sutton-Smith (1967) to speculate that creativity and
playfulness were related because they are both aspects
of intelligence. Indeed, when Singer & Rummo (1973)
reanalyzed Lieberman's (1965) data, they found the
relationship between playfulness and creativity was

dependent on mental age. It was, therefore, expected

38

in the current study that playfulness and creativity
would be related, but not when mental age was partialled

out.

Hypothesis 5

Playfulness should be related to creativity.

Hypothesis 6

 

When mental age is partialled out, playfulness
and creativity should be unrelated.

CHAPTER II

METHOD

Subjects

The subjects were 90 kindergarten-aged children
(45 boys, 45 girls) from the greater Lansing area. They
ranged in age from 5.33 years to 7.17 years (median =
6.00 years). They all appeared to be from white middle-

class families.

Measures

Four tests were administered to all children and
a rating scale was used to describe children in one of
the conditions.

The Alternate Uses Test (Wallach & Kogan, 1965)
was used to provide comparability with Dansky and
Silverman's (1973, 1975) studies. The Alternate Uses
Test measures associative fluency: the child is asked
to name all the uses for an object that he/she can
think of. Four of these questions were asked. The test
was not timed, but took approximately 10 minutes.

The Barron-Welsh Art Scale (Barron & Welsh,
1963), found to be Opposite of associative fluency in

the pilot study, consists of 86 drawings. Since artists

39

40

were found to like complex, asymmetrical drawings and

to dislike simple, symmetrical drawings, the closer a
person follows this pattern the more creative that

person is considered. Although primarily used with
adults, it has been administered to children (Ward, 1968).
The test was untimed, but took approximately five

minutes.

Figural flexibility was found to be independent
of associative fluency in the pilot study. Figural
flexibility is calculated from the Picture Completion
and Lines parts of the Torrance Figural Test (Torrance,
1966b). However, each part is timed and takes 10
minutes. To reduce the testing time somewhat it was
decided to use only one part of the Torrance Figural
Test. Since in the pilot study the Flexibility score
from the Lines part correlated better with the total
Flexibility score (r = .9119) than did the Flexibility
score from the Picture Completion part, only the Lines
part was used.

The Peabody Picture Vocabulary Test is a measure
of intelligence that is easily administered to kinder-
garten children. In the test a series of pictures are
placed before the child. The tester reads a word and
the child points to the picture out of 4 that corres-
ponds with that word. The test takes a short time

(approximately 10 minutes) and has been used in other

41

studis of play and creativity (e.g., J. E. Johnson,
1976; Lieberman, 1965).

The Playfulness Scale (Lieberman, 1977) was used
to rate the play of the children. The scale measures
five variables: physical spontaneity, manifest joy,
sense of humor, social spontaneity, and cognitive spon-
taneity. Each child is rated for the quality and
quantity of each variable. There are also ratings for
such "dummy" variables as intelligence and attractive-
ness. The scale was adapted in this study, since
children were tested individually and thus social spon-

taneity could not be rated.

Materials

 

For experimental conditions several toys were
used. They included: approximately 30 paper clips,
15 blank 3 x 5 cards, 10 empty matchboxes, small corks,
wooden spools, pipe cleaners, plastic clothes pins,
pliers, paper towels, and paper cups. These toys were
selected because they are not objects children typically
play with, thereby fostering creativity, and because they
have been used in previous studies (Dansky & Silverman,
1973, 1975). In addition a box of crayons was available

for use by the child in various parts of the study.

42

Procedure

 

Part 1

Approximately one week before the testing began,
the experimenter visited the classroom of the children
to be tested for an hour to build up rapport. When the
testing began, each child was randomly assigned to one
of the three conditions. Thus 15 girls and 15 boys were
assigned to each condition.

Each child was brought individually to the test-
ing area. In the testing area there were a table with
two chairs opposite each other and a divider behind
which observers could sit and not be seen. On the
table there were: paper clips, blank 3 x 5 cards, empty
matchboxes, small corks, wooden spools, pipe cleaners,

plastic clothes pins, and pliers.

Play Condition. The child was seated at the

 

table and given the following instructions:

I like to see what kinds of games children
like to play. I like to help children to have
fun. Today I brought the things that you see here
on the table for you to play with. They may not
look like toys, but I have seen that children can
have a lot of fun playing with them. I would just
like to let you have some fun and play with these
things any way that you would like. Do whatever
you want to do with them.

The child was then allowed to play for 10 minutes.

This 10 minute period rised in this and other conditions

43

was used to provide comparability with previous work
(Dansky & Silverman, 1973, 1975).

Imitation Condition. The child was seated at

 

the table and given the following instructions:

I have some things for you to do today. Right
now I would like you to watch me to see what I do.
Then I will ask you to do what I have done. Remem-
ber, watch me carefully so that you will be able
to do just what I do.

The experimenter demonstrated four tasks to the
child which the child imitated after the experimenter
demonstrated them. These tasks were: (1) attaching
paper clips to a 3 x 5 card; (2) putting a pipe cleaner
through the hole in a spool and twisting the ends
together; (3) opening empty matchboxes and using the
pliers to fill them with corks; and (4) clipping
clothes pins end to end to form abstract configurations.
The order of these tasks was varied. Each task took

2 1/2 minutes.

Control Condition. The child was seated at

 

the table (the toys absent) with a box of crayons and

four sketches. The child was then told to color these
pictures in any way the child desired. The child was

allowed to color for 10 minutes.

Post-Tests. After the treatment session, the

 

child was given the three creativity tests (the Alter-
nate Uses Test, the Barron-Welsh Art Scale, and the Lines

part of the Torrance Figural Test) and the Peabody

44

Picture Vocabulary Test. The order of the creativity
tests was randomized so that each test was first, second,
and third an equal number of times. Thus five boys and
five girls within each condition were assigned to each
of the three orders. The Peabody Picture Vocabulary Test
was always presented last.

During the Alternate Uses Test the children were
asked to name all the uses for objects they may have
had contact with (paper clip and matchbox) and those
they had not (paper towel and paper cup). In all cases
the children were permitted to see, but not handle the
objects.

During the pilot study it was found to be quicker
and easier to administer the Barron-Welsch Art Scale
by pointing to each of the pictures and asking the
children whether they liked or disliked the picture. The
same procedure was followed here.

Administration of the Lines part of the Torrance
Figural Test followed standard form. Children were
asked to draw as many different kinds of pictures or
objects as possible from the 30 pairs of lines in 10

minutes.

Part 2
During the play condition two judges were seated

behind the divider from which they could observe the

45

child's behavior unseen. (In the other conditions the
divider was in place without the presence of observers.)
Once the child was told to play, the judges indicated

on a prepared sheet (see Figure l) at 15 second inter—
vals noted by a beeper which of the eight toys the child
made use of. If in that interval the child made use of
more than one object, the judge noted whether that use
had been separate (i.e., the child moved from contact
with one toy to another without any connection between
the two) or integrated (i.e., the toys were used togetherh
Use of only one toy in an interval was, by definition,
separate. Further, each judge had to decide whether the
child's contact with the toy(s) was exploration (i.e.,
the child showed some uncertainty; affect was neutral or
mildly negative; and the child's actions toward the
toy(s) could have included: scanning the set, focusing
on one toy; examining the toy(s) visually and physically),
functional play (i.e., no uncertainty, positive affect,
activity done for its own sake but bounded by the
physical aspects of the toy(s)),or imaginative play
(i.e., the same as functional play, except the activity
involves pretending). (The above distinctions corres-
pond to those noted by Singer [1973] and Weisler and

McCall [1976].)

46

Figure 1
Recording Sheet

Tile Paper 325 Match Clothes Fun Ina
Inter Clips Cards Boxes Corks Spools Cleaners Pins Pliers Into; Sap Explo Play Play

1 o . ‘ 3
o . . 7 ‘ I

 

47

After the play session each judge rated the
child on the adapted version of the Playfulness Scale
(see Figure 2).

As can be seen in Tables 1 and 2, interjudge
reliabilities were quite high. Table 1 presents inter-
judge agreements on the objects and descriptions of the
play behavior. In all cases percentages of agreement
are over 95%, and in all cases but one reliabilities
are over .90.

In Table 2 interjudge reliabilities on the Play-
fulness Scale are presented. In the first column are
those reliabilities from Lieberman's (1964) study, in
the second column are those from the current study. It
can be seen by examining the two columns that the
reliabilities are comparable. This is somewhat sur-
prising given the amount of time spent observing the
children in the two studies. In Lieberman's (1964)
study the judges were two teachers who had observed
the children over quite a period of time, while in
the current study the two judges had observed the

children for only 10 minutes!

I. A.
B.
II. A.
B.

Figure 2.

48

How often does the child engage in spontaneous physical
movement and activity during play?

This behavior includes hand clapping, moving in seat,
turning head from side, and other movements of the whole
body or parts of the body that could be judged as a fairly
clear indication of exuberance.

Very Very
Often Often Occasionally Rarely Rarely
5 4 3 2 1

How is his/her motor coordination during physical activity?

Excellent Very Good Good Fair Poor

5 4 3 2 1

How often does the child show joy in or during his/her
play activities?

This may be judged by facial expression, such as smiling,
by verbal expressions, such as saying "I like this,"
more indirect vocalizing, such as singing as an accom-
paniment of the activity, or by repetition or resumption
of activity with clear evidence of enjoyment.

Very Very
Often Often Occasionally Rarely Rarely
5 4 3 2 1

With what freedom of expression does he/she show joy?

This may be judged by the intensity of loudness of a
chuckle or sing-song, as well as by the child's ability

to repeat or resume his/her activity by his/her own choice.

 

Very
High High Moderate Some Little
5 4 3 2 1

Adapted Version of Playfulness Scale (Lieberman, 1977)

49

Figure 2 Continued

III.

IV.

A.

A.

How often does the child show a sense of humor during play?
By "sense of humor" is meant rhyming and "glint-in-the-
eye" behavior, as well as the ability to see a situation
as funny as it pertains to himself/herself or others.

Very Very
Often Often Occasionally Rarely Rarely
5 4 3 2 1

With what degree of consistency is humor shown? This may
be judged by its occurence across situations.

Very
High High Moderate Some Little
5 4 3 2 1

How often does the child show spontaneity during expres-
sive and dramatic play? ‘

Instances of such behavior include labeling objects with
names other than commonly used, or changing those names in
different situations, especially those with living beings.

 

Very Very
Often Often Occasionally Rarely Rarely
5 4 3 2 1

What degree of imagination does the child show in his/her
expressive dramatic play?

Instances of imagination include using toys for other than
accepted usage, as well as incorporating nonexistent
objects into the play situation.

Very
High High Moderate Some Little
5 4 3 2 1

How bright is the child?
This is your estimate of the child's intelligence based on
observed behavior or inferred potential.

Extremely Moderately Not Too
Bright Bright Average Bright Bright

5 4 3 2 l

50

Figure 2 Continued

VI. How attractive is the child?
This is your evaluation of the child's physical appeal.

Passable in Somewhat
Beautiful Attractive Looking Looks and homely and
Appearance Unattractive

 

51

Table 1

Reliability of Judges on P1ay--Objects
and Descriptions

 

 

Percent
Agreement Reliabilitya
Paper Clips 97 .95
3 x 5 Cards 99 .94
Match Boxes 98 .97
Corks 96 .94
Spools 97 .96
Pipe Cleaners 98 .98
Clothes Pins 96 .96
Pliers 99 .96
Integrated-Separate 96 .91
Exploratory-Functional- 97 .89

Imaginative

 

aSpearman-Brown prophecy formula

52

Table 2

Reliability of Playfulness Scale

 

 

Lieberman Current
(1964) Study
Physical Spontaneity .Aa .83 .80
B .67 .76
Manifest Joy A .73 .72
B .70 .74
Sense of Humor A .66 .63
B .70 .63
Social Spontaneity A .74 -
B .67 -
Cognitive Spontaneity A .68 .82
B .67 .75
Intelligence .74 .64
Physical Attractiveness .72 .45
b
Total Scale .94 -
Total Scale (Without Social .93 .86
Spontaneity)
Total Scale (Without Intelli- .96 -
gence and Physical Attrac-
tiveness)
Total Scale (Without Social .95 .86

Spontaneity, Intelligence,

and Phy

sical Attractiveness)

 

aSpearman-Brown prophecy formula

b

Cronbach's alpha

CHAPTER III

RESULTS

The Effect of Play on Creativity:
Experimental Analysis

 

 

Seven creativity scores were calculated from the
three creativity tests: number (AUN) and uniqueness
(AUU) from the Alternate Uses Test (wallach & Kogan, 1965»
the Revised Art Scale score from the Barron-Welsch Art
Scale (BWAS) (Barron & Welsh, 1963); and fluency (FLU),
flexibility (FLEX), originality (ORIG), and elaboration
(ELAB) scores from the Lines part of the Torrance
Figural Test (Torrance, 1966b).

These scores were analyzed in a 3 (Treatment) x
3 (Order) x 2 (Sex) multivariate analysis of variance
(Finn, 1974). The results from this analysis are dis-
played in Tables 3 to 9.

In Hypothesis 1 it was expected that play would
aid performance on the AUN and the AUU, hinder perform-
ance on the BWAS, and have no effect on the FLU, FLEX,
ORIG, and ELAB. As can be seen in Table 3 the multi-
variate F for the effects of treatment was not signifi-
cant. In examining the effects on specific scores, alpha
should be reduced to .007 to account for the multiple

measures. As can be seen in Table 3, none of the

53

54

Table 3
Multivariate Analysis of Variance for the
Effects of Treatment on Measures
of Creativity

Multivariate F = 1.2493 df = 14,132 p < .2480

 

 

Variable Univariate F p Stepdown F p

AUN .4556 .6360 .4556 .6360
AUU .2784 .7579 .2188 .8040
BWAS .2266 .7979 .3097 .7347
FLU 1.4238 .2476 1.5899 .2114
FLEX 5.0225 .0091 3.8586 .0259
ORIG 3.6855 .0300 2.2689 .1114
ELAB .2511 .7787 .1074 .8984

 

Univariate df = 2,72

Table 4
Multivariate Analysis of Variance for the
Effects of Order on Measures
of Creativity

Multivariate F = .4094 df = 14,132 p < .9699

 

 

Variable Univariate F P Stepdown F P

AUN .1804 .8354 .1804 .8354
AUU .2750 .7604 .1755 .8394
BWAS .8940 .4135 .8420 .4352
FLU .3759 .6881 .3087 .7355
FLEX .5198 .5969 .6450 .5279
ORIG .9652 .3858 .5672 .5698
ELAB .6009 .5511 .2259 .7984

 

Univariate df = 2,72

55

Table 5
Multivariate Analysis of Variance for the
Effects of Sex on Measures
of Creativity

Multivariate F = 1.3217 df = 7,66 p < .2542

 

 

Variable Univariate F p Stepdown F p

AUN .6680 .4165 .6680 .4165
AUU 1.4599 .2309 1.7274 .1930
BWAS 1.5014 .2245 1.6640 .2014
FLU .0668 .7969 .0002 .9894
FLEX 2.1630 .1458 3.5613 .0635
ORIG .2036 .6533 1.2007 .2772
ELAB .0359 .8504 .3710 .5446

 

Univariate df = 2,72

Table 6
Multivariate Analysis of Variance for the
Effects of Sex by Order Interaction
of Creativity

Multivariate F = .8521 df = 14,132 p < .6119

 

 

Variable Univariate p Stepdown F p

AUN 1.8600 .1631 1.8600 .1631
AUU 1.4766 .2353 .3937 .6761
BWAS .0617 .9403 .0647 .9374
FLU .1625 .8504 .0389 .9619
FLEX .3311 .7193 .4136 .6630
ORIG .0503 .9510 .0473 .9539
ELAB 2.4036 .0977 3.2945 .0433

 

Univariate df = 2,72

56

Table 7

Multivariate Analysis of Variance for the
Effects of Treatment by Sex Interaction
on Measures of Creativity

Multivariate F = 1.1103 df = 14,132 p < .3547

 

 

Variable Univariate F p Stepdown F p

AUN 2.6663 .0764 2.6663 .0764
AUU 3.1849 .0473 .5004 .6085
BWAS 1.4235 .2476 1.3912 .2556
FLU 1.5230 .2251 1.4488 .2420
FLEX .7665 .4684 .5552 .5766
ORIG 1.0570 .3529 .4217 .6577
ELAB .4664 .6292 .9502 .3919

 

Univariate df = 2,72

Table 8

Multivariate Analysis of Variance for the
Effects of Treatment by Order Interaction
on Measures of Creativity

Multivariate F = 1.3355 df = 28,239.3886 p < .1282

 

 

Variable Univariate F p Stepdown F p

AUN 2.0298 .0993 2.0298 .0993
AUU 2.1896 .0787 .5087 .7295
BWAS .7004 .5943 .7471 .5633
FLU .9339 .4494 .6778 .6097
FLEX 1.8831 .1227 2.1729 .0813
ORIG .3408 .8496 1.1159 .3565
ELAB 1.7040 .1585 2.1613 .0813

 

Univariate df = 4,72

57

Table 9

Multivariate Analysis of Variance for the
Effects of Treatment by Sex by Order
Interaction on Measures of
Creativity

Multivariate F = 1.0538 df = 28,239.3886 p < .3972

 

 

Variabe Univariate F p Stepdown F p

AUN .4382 .7806 .4382 .7806
AUU .5501 .6996 .2115 .9313
BWAS .3863 .4767 .9455 .4431
FLU .1399 .9699 .1318 .9703
FLEX .3085 .8714 .6926 .5997
ORIG 1.3786 .2500 3.4080 .0135
ELAB 2.3498 .0623 1.6957 .1615

 

Univariate df = 4,72

58

univariate Fis nor stepdown F's reach this level,
although effects for flexibility (univariate F = 5.0225,
df = 2,72, p < .0091) and originality (univariate F =
3.6855, df = 2,72, p < .0300) approached it. When the
means for the groups were examined, it was found that
children in the imitation condition performed better
than those in the play and control conditions for both
flexibility (I = 7.83, P = 4.77, C = 5.43) and origin-
ality (I = 17.33, P = 11.37, C = 10.37).

Further information about Hypothesis 1 pertaining
to the effect of play on performance on the Alternate
Uses Test can be seen by examining Table 7. There is a
near significant treatment by sex effect for uniqueness
(univariate F = 3.1849, df = 2,72, p < .0473) and
number (univariate F = 2.6663, df = 2,72, p < .0764).
Examination of the means revealed that the expectation
in Hypothesis 1 that children in the play condition
would do better on the Alternate Uses Test than those
in the other two conditions was true for the girls
(uniqueness: P = 8.47, I = 5.60, C = 2.93; number:

P = 16.33, I 12.93, C = 9.87) but not for the boys

(uniqueness: P = 6.20, I = 6.47, C = 9.73; number:
P = 13.47, I = 13.33, C = 16.53).

Examination of Tables 4, 5, 6, and 9 reveal
little of interest with regard to the effects of order

(Table 4), sex (Table 5), sex by order interaction

59

(Table 6), and treatment by sex by order interaction
(Table 9), There are some interesting effects of the
treatment by order interaction (Table 8), which were
not hypothesized.

In Table 8 there are near significant effects
of the treatment by order interaction for the number
(univariate F = 2.0298, df = 4,72, p < .0993) and
uniqueness measures (univariate F = 2.1896, df = 4,72,
p < .0787) of the Alternate Uses Test. Examination of
the means revealed that children in the play condition
did best on the Alternate Uses Test when it was given
second (uniqueness: l = 6.3, 2 = 11.6, 3 = 4.1; number:
1 = 13.7, 2 = 19.7, 3 = 11.3), while those in the
imitation and control conditions did best when it was
given third--imitation (uniqueness: 1 = 6.0, 2 = 5.7,
3 = 6.4; number: 1 = 13.1, 2 = 12.1, 3 = 14.2), control
(uniqueness: l = 5.2, 2 = 4.2, 3 = 9.6; number: 1 =

12.3, 2 = 10.9, 3 = 16.4).
To test Hypothesis 2 that effects of mental age

would not effect the predictions of Hypothesis 1, a
multivariate analysis of covariance was carried out on
the same data with mental age, calculated from the
Peabody Picture Vocabulary Test, as a covariate. Since
mental age did not interact with the independent vari-
ables (F = .3236, df = 7,65, p < .9407), it was legiti-

mate to attempt this analysis. The results were much

60

the same as in Tables 3 to 9. This can be seen by
comparing the multivariate F's in Table 10 with those

in Tables 3 to 9.

Measures Derived From Observations

of Play

The observer's judgments on each of the 30
children in the play condition were coded on the scoring
sheet shown in Figure 1. For each of the 40 intervals
the following scoring system was used: If the judges
agreed that an object was not played with or a category
was inappropriate, a zero was scored. If they agreed
that an object was played with or a category was
appropriate, a two was scored. If only one judge
thought an object was played with or a category was
appropriate, a one was scored. In this way there were
520 scores (40 intervals by 8 objects and 5 categories)
for each child. (It should be noted that one child was
inadvertently permitted to play for only 9 minutes and
thus had 468 scores.)

A number of variables were created from these
scores. First, for each of the 8 objects and 5 cate-
gories scores were summed across the 40 intervals to
give total object and category use. (It should be noted
that the integrated and separate categories, as well as
the exploratory, functional and imaginative categories,

are ipsative measures and total to 80.)

61

Table 10

Multivariate Analysis of Covariance for the
Effects of Treatment, Order, and Sex
of Creativity

 

 

Multivariate
Effect F df p
Treatment 1.2321 14,130 .2600
Order .4005 14,130 .9727
Sex 1.3124 7,65 .2588
Sex by Order .8884 14,130 .5729
Treatment by 1.0977 14,130 .3658
Sex
Treatment by 1.3017 28,235.7830 .1499
Order
Treatment by 1.0247 28,235.7830 .4362

Sex by Order

 

62

The complexity of each child's play, or the
number of objects being played with at a given period
of time, was thought to be an important consideration
and was calculated in the following way: Within each
of the 40 intervals the scores for the 8 objects were
summed. It was felt that complex play would only occur
if the use of those objects were integrated. So the
sum for the 8 objects was then multiplied by the inte-
grated score. These 40 products were then summed to
form one total score for complexity.

Also of interest were the shifts the child
showed in the play, or the number of times a child
added or subtracted an object from the set being played
with. Shifts were examined in two ways. The first way
involved looking at the pattern of scores for the 8
objects and 5 categories within each interval. These
patterns were compared in the following manner: If the
pattern from one interval to the next were the same,

a zero was scored. If the two patterns were different,
a one was scored. The total of these scores was called
global shifts.

It was felt that this approach to shifts might
miss some character of them. For example, the global
shifts score would be the same whether one object or
more were added to the set. To remove this problem a

second shifts score was calculated. The score of each

63

object was compared from one interval to the next. If
the scores were the same, a zero was scored. If the
two scores were different, a one was scored. These
scores were then summed for each object. These scores
in turn were summed to give an object shifts score.
The Relationship Between Measures of
Creativity,—Playfulness, andPlay

To examine Hypothesis 3 that time spent playing
with an object would be unrelated to the number and
uniqueness of uses given for that object, separate
scores were calculated for the four objects used in the
Alternate Uses Test. Thus there were number and unique-
ness scores for the objects played with (paper clip and
matchbox) as well as for those that were not (paper
towel and paper cup).

The following variables were then correlated
(abbreviations are in parentheses): sex, mental age
(MA), number and uniqueness of uses for the paper clip
(N1 and U1), number and uniqueness of uses for the paper
towel (N2 and U2), number and uniqueness of uses for
the matchbox (N3 and U3), number and uniqueness of uses
for the paper cup (N4 and U4), the total number and
uniqueness scores for the Alternate Uses Test (AUN and
AUU), the Revised Art Scale score from the Barron—Welsh
Art Scale (BWAS), the fluency, flexibility, originality,

and elaboration scores from the Lines part of the

64

Torrance Figural Test (FLU, FLEX, ORIG, and ELAB), both
parts of the Physical Spontaneity, Manifest Joy, Sense
of Humor, and Cognitive Spontaneity items as well as
Intelligence and Physical Attractiveness items from the
Playfulness Scale (P81, P82, JOYl, JOY2,-SH1, 8H2, C51,
C52, INT, and PA), object use totals for the paper clips,
3 x 5 cards, matchboxes, corks, spools, pipe cleaners,
clothes pins, and pliers (PC, CARD, MB, CRK, SP, CL,
CP, and PL), category use for integrated, separate,
exploratory, functional, and imaginative play (INTEG,
SEP, EXP, FUN, and IMAG), complexity (COMPEX), and
global and object shifts (GSHIFT and OSHIFT). The
correlations are presented in Appendix A.

Examination of Table 11 indicates that Hypothesis
3, that time spent playing with an object should be
unrelated to the number and uniqueness of uses given
for that object, was supported. Correlations between
use of paper clips in play and the number and uniqueness
of uses for them are .18 and .14 respectively,both not
significant. Similarly, use of the matchboxes corre—
lated .09 and .03 with number and uniqueness of uses
for them respectively.

Some other points should be noted from the
correlations in Appendix A. Only three variables corre-
late significantly with sex (positive correlations mean

higher scores by boys, negative correlations higher

65

Table 11

Selected Correlations Between Measures
of Creativity, Playfulness, and Play

 

Correlations Between Time Spent Playing With
Objects and Alternate Uses For Them

 

 

 

 

 

N1 U1 N3 U3

PC .18 .14 MB .09 .03
Correlations Between Measures of Play
and Creativity
INTEG IMAG COMPLEX OSHIFT GSHIFT

N1 .04 .19 .11 -.03 -.O3
Ul -.03 .35 .04 .04 .08
N2 .00 .32 -.01 -.17 -.13
U2 .12 .23 .11 -.08 -.07
N3 -.23 .33 -.24 -.09 -.04
U3 -.20 .15 -.19 -.12 -.10
N4 .12 .32 .16 .14 .15
U4 .12 .37 .16 .13 .14
AUN .00 .31 .03 —.06 -.05
AUU .01 .34 .06 .00 .03
BWAS .36 -.15 .33 -.03 -.09
FLU .08 .01 -.08 .02 .10
FLEX —.06 -.02 -.09 .03 .03
ORIG .06 .03 -.08 -.07 -.06
ELAB -.01 .37 -.07 —.08 -.04

 

66

scores for girls): physical spontaneity--both parts--
and the second part of cognitive spontaneity. Three
significant correlations out of 42 are not much more
than that expected by chance. As a result, there is
no further information why the associative fluency
effect was found only for girls.

The other intriguing finding is that the highest
correlation of mental age with any other variable is
.59 with the estimate of intelligence from the Playful-
ness Scale. This means that the two observers were
able to make fairly accurate judgments of the children's
mental age from 10 minutes observation of their play.

Examination of Table 11 provides information on
the status of Hypothesis 4, that complexity, integra-
tion, and imaginativeness of play should be related to
creativity. For complexity the hypothesis was not
verified. The highest correlation for complexity and
a creativity measure is with the Barron-Welsh Art Scale,
.33. The measures of shifts, global and object, fare
just as badly.

Integration does little better, reaching signi-
ficance only with the Barron—Welsh Art Scale, .36.
Imaginativeness does somewhat better, correlating near

significance with most of the Alternate Uses Test items

67

(correlations range from .15 to .37) and at significance
with the elaboration score of the Torrance Figural Test,
.37.

Multidimensional Scaling and Clustering

of Creativity, Playfulness, and Play

One purpose of the current study was to suggest
a model which related certain aspects of play with
certain kinds of creativity. To do this the correlation
matrix, minus the scores for the individual objects on
the Alternate Uses Test, was submitted to multidimen-
sional scaling (Kruskal, 1964). A two dimensional
solution with a stress of .254, a reasonable value given
the large number of variables, was generated. This
solution is presented in Figure 3.

The two dimensions are interpretable. The first
dimension is a continuum from separate-imaginative play
to a more concrete integrative-functional play. The
second dimension appears to be a cognitive style dimen-
sion, ranging from creativity through intelligence to
shifts (or impulsivity to use Kagan, Rosman, Day,
Albert, and Phillips'll964] term).

It was thought that clustering these variables
would be useful in proposing a model of the relationship
between play and creativity. The clustering solution,
using S. C. Johnson's (1967) special method, is pre-

sented in Figure 4. It was then applied to the scaling

68

32— has one
.nuoanauhuum $339695 me .8336» magnum 19333»?!

a one:

 

E88 8:8
on E38
use
a.
an. an
E2.
.5
«we
«you :5
3..
Se 8
EH 8
a:
a 33 §
«mm
a B

 

69

38 5.. x.- ..uoflstfim 51386 «o 3.5.38
.4 953»

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

70

solution in the following way. Variables were circled
as they clustered hierarchically. Clustering continued
as long as the clusters made sense statistically and
logically, e.g., verbal and nonverbal creativity were
not allowed to cluster together. The resulting solution
is presented in Figure 5.

As can be seen, especially in Figure 5, the
multidimensional scaling solution does not suggest a
model for the relationship between play and creativity.
The creativity, playfulness, and play variables all
cluster separately. Thus no comparison was made with
Lieberman's (1964) data.

Path Analytic Model of Playfulness,
Play, and Creativity

 

 

While the multidimensional scaling solution did
not provide a model for the relationship between play
and creativity, the clustering solution did suggest
that certain variables could be grouped together. The
six final clusters from Figure 5 are presented and
described in Table 12.

The six clusters were meaningful and internally
consistent (alphas greater than .79). The clusters,
verbal and nonverbal creativity, should be self-evident
from previous discussions of the Alternate Uses Test

and the Torrance Figural Test. Use of the 3 x 5 cards,

71

 

 

88 SE is 63238: 53388
new 3336 as. .6338 838m 33333.:

n 33:

 

 

 

Table 12

Clusters Formed from the Creativity,
Playfulness, and Play Variables

 

Verbal Creativity (VC)

Alternate Uses--Number
Alternate Uses--Uniqueness

Shifts (SH)

3 x 5 Card Use
Matchbox Use
Cork Use
Global Shifts
Object Shifts

Playfulness-Intelligence
(PI)

Mental Age

Physical Spontaneity--Part 1
Physical Spontaneity--Part 2

Cognitive Spontaneity--
Part 2
Intelligence

Nonverbal Creativity (NVC)

Fluency
Flexibility
Originality

Complexity (CX)

Paper Clip Use
Spool Use

Pipe Cleaner Use
Integration
Complexity

Playfulness-~Fun (PF)

Joy--Part l
Joy--Part 2
Sense of Humor--Part 1
Sense of Humor--Part 2

Cognitive Spontaneity--
Part 1
Imaginativeness

 

73

matchboxes, and corks were included in the shifts
cluster, since these objects were easy to add or sub-
tract from the set being played with. Use of paper
clips, spools, and pipe cleaners were included in the
complexity cluster, since these objects were frequently
used in combination with others. The Playfulness Scale
with a few other items was divided into two parts:

one dealing with intelligence in play, the other with
the fun aspect of play. This distinction is consistent
with dual aspects of cognition and affect in Lieberman's
(1977) notion of playfulness and with Singer's (1967)
distinction between reminiscent and playful daydreams.

These clusters were then correlated using Hunter
and Cohen's (Note 2) PACKAGE program. The oblique
multiple groups analysis routine in this program was
performed with the six clusters as the groups. The
analysis standardized all the variables that were
part of the groups and gave them equal weighting within
the groups. The groups were then correlated, correcting
for attenuation (see Table 13).

These correlations were submitted to path
analysis using the model presented in Figure 6. This
model proposes that playfulness, a personality con-
struct, should be evident in play, which in turn should

effect creativity. While it can be argued that since

Correlations Corrected for Attenuation
Between Clusters of Creativity,

74

Table 13

Playfulness, and Play

 

 

pr p1 sa cx vc
PI .61

SH .08 .57

cx -.09 .45 .34

vc .13 .16 -.03 -.01

75

e358 use: .8533: .2 .822 3223 sea 2:

 

 

 

 

 

 

 

 

 

m 959...
A>._..>F<wmo (flirt---) “taxman. .200 T. ....... A mozwwjuwkz:
A .Emmmi , x . 1., \Emmzétfie
Va... VA.
'\\\\ III! F\\ III/
A >._._>_._.<mm8: .......... A ”Kim 7-2:... [A 2:... v

 

3<mmw>zozv Tmmwzqambjdv

 

 

76

the ratings of playfulness followed the judges' obser-
vations of play, the play clusters should precede the
playfulness clusters in the path model, this model was
chosen because it more closely matched the relationship
among playfulness, play, and creativity proposed by
Lieberman (1977).

The fit between the observed and predicted
correlations was very good--the sum of the squared
deviations was .16 (see Table 14). The path model with
the ordinary least squares estimates of the path coeffi-
cients is presented in Figure 7. The path coefficients
are calculated by doing a regression of each cluster
onto its causal antecedents. With two or more antece-
dents, as in this model, the path coefficients are the
beta weights. The value of the double curved arrow
between playfulness-fun and playfulness-intelligence is
simply the correlation between them.

As can be seen in Figure 7, playfulness-fun and
playfulness-intelligence have very different impacts on
shifts and complexity. It can be noted by examining
Table 13 that the negative path coefficients from play-
fulness-fun in Figure 7 are not due to a suppressor
effect. Children who are high on playfulness-fun (i.e.,
who outwardly express joy during play) are likely to
play with a small set objects and make few changes in

which objects belong to the set. Children who are high

77

Table 14

The Basic Assessment of the Path Analysis
of Playfulness, Play, and Creativity

 

Observed Correlations

 

 

 

 

 

 

 

 

PF PI SH CX VC NVC
PF 1.00 .61 .08 -.09 .13 .18
PI .61 1.00 .57 .45 .16 .14
SH .08 .57 1.00 .34 —.03 .06
CX -.09 .45 .34 1.00 -.01 .01
VC .13 .16 -.03 -.01 1.00 .26
NVC .18 .14 .06 .01 .26 1.00

Predicted Correlations

PF PI SH CX VC NVC
PF 1.00 .61 .08 -.09 .00 .01
PI .61 1.00 .57 .45 -.02 .03
SH .08 .57 1.00 .41 -.03 .06
CX -.09 .45 .41 1.00 -.01 .01
VC .00 -.02 -.03 -.01 1.00 .00
NVC .01 .03 .06 .01 .00 1.00

Observed Minus Predicted Correlations

PF PI SH CX VC NVC
PF .00 .00 .00 .00 .13 .17
PI .00 .00 .00 .00 .18 .11
SH .00 .00 .00 -.07 .00 .00
CX .00 .00 -.07 .00 .00 .00
VC .13 .18 .00 .00 .00 .26

 

356530 58 be meBEzmm 88:8 .30.. res—:0 e5
55 £2.85 Posse £2.23: es 382 2584 sea 2:

 

 

 

 

 

 

 

 

 

N. 059....

00. em.
itasmmo 3.5)-..-«1 >txmdzoo T ....... A mozmsjmez:
l .2633 . m, \..mmmz.5h_>5$

n _o .. ....mm.

\t/ \A/ .
\\\\ IIIMOV K\\ IIIBV _m
;t>_2wmo.».:::::wl mtim 7.5-4-1 22.:

 

34mmm>zozo me. me- l-mmmz._:h_><._&

 

 

79

on playfulness-intelligence (i.e., who view play as an
intellectual challenge) are likely to play with a large
set of objects and make many changes in which objects
belong to that set.

The other point of note in Figure 7 is the near
zero path coefficents between shifts and complexity,
and verbal and nonverbal creativity. This should indicate
that play, at least as defined by these clusters, has
a weak effect on creativity.

Redundancy Analysis of Playfulness
and Creativity

 

 

The final questions to be resolved were Hypo-
theses 5 and 6, that playfulness is related to creativity,
but not when mental age is partialled. To test these
hypotheses, playfulness was defined as the Playfulness
Scale items, except for the Physical Attractiveness
and Intelligence items, which Lieberman (1977) has
termed "dummy" items. Creativity was defined as the
AUN, AUU, BWAS, FLU, FLEX, ORIG, and ELAB. The corre-
lations among these variables, which can also be found
in Appendix A, are presented in Table 15.

To test Hypothesis 5 it was decided to run a
redunancy analysis between the two sets of variables
(Van den WOllenberg, 1977). Redundancy analysis com-
putes the variance accounted for in one set of variables

by another set of variables. It was chosen instead of

80

 

 

oo.H ho. co. mo.l mH.I «H. vH. hN. mm. on. mN. 5H. 0H. 5H. «N. mijm
co...” me. an. mo. NN. ON. CO. NO... ON. 50. AN. NH. mN. ON. me0
cotn .3. mo.l 3”. ma. oo. mo... MN. 3...! MN. .8.I mN. oo. xmqm
oo.H mo.l MN. 0N. vo. mo.l mo. mo. ma. HN. oe. MN. HeHm
co...” No.l No. co. on mof «N... Nm.I mm.l ON. OH... 955
ooé mm. ma. 0N. 5N. NH.I mo. No. mo. mo. DDfl
004 NH. ma. mN. mo.l mo. 00. oo. Ho.l 754
cc...” mm. mN. Nm. mN. «v. mm. N5. NmU
004 «v. «m. mm. Nw. NH. «m. HmU
oo.H mm. mm. hm. ma. «N. Nmm
coin mm. 2.. ha. 5m. .26
cot... mm. ma. mm. NNOU
ooé 5H. vm. HMOD
oo.H mm. Nmm
oo.H .Hmm
mfifim QURV thm .52 “TEE n5d 22m .dmw QRK..GHE Nmm Hfim

NmU H8 Nam

 

sue>nnmmuo

moccasmamHm

 

hawbflumouo can mumsH5m>mam mo monomcmz macE¢ mcowucamnuou

ma manna

81

canonical correlation, because canonical correlation can,
in maximizing the correlation between two sets of vari-
ables, weight heavily two minor components, one from
each set, which are highly correlated, but explain
little of the shared variance between the sets.

Thissen and Wollenberg (Note 3) provided me
with the program for redunancy analysis, REDANAL, which
I adapted for current use on the computer. The results
of the first analysis are presented in Tables 16 and
17.

Table 16 presents the factor loadings of the
playfulness and creativity variables on the redundancy
variates. From these loadings it can be determined that
the redundancy variates of playfulness explain 31.8%
of the variance of creativity, while the redundancy
variates of creativity explain 22.2% of the variance of
playfulness. (Redundancy analysis is not symmetric.)
It can be said that playfulness and creativity have a
good deal in common.

This can also be seen in Table 17. The unro-
tated correlations show high correlations between a
number of the variates. (This also illustrates the
lack of bi-orthogonality in redundancy analysis.) The
correlations among the rotated redundancy variates,

which are roughly equivalent to those that would be

82

 

 

MN.I «0. NM. MN.I MN... MH... Mb... oo. mo. mof MH... MM. o«.l MH... gm
«M. Nm.l MH... ow. bH.I «o. m«.l 00. «H... No. Mo.l 0N. MH. H«.I me0
OH... MN. MM... Hh. om.l mo. MH.I Ho. No. hH.I no. MH.I «N. mm.l xmnm
mm. HH. mH... hm. Mo. 5M. o«.l Ho.l mo. NH. mo... MH. 5M. o«.l DHm
«H. hN.I N«. Ho... om... mm. H«. oo. Ho.l No. mm... MN... NH... ON... 955
pm. MH. «M.I MN... M«.I ho... MN.I mo. Ho. mo. «H. hof «M... H«.I Dam
Mb. «N. MH.I HN.I m«.l 0N.I MN.I wof oo. «0. HH. mof MM.I MM.I 254
M0... «0. M0... 0N.I HH... MH. hN... o«. mm. Ho. Ho.l MN. NM... MH... Nmo
Ho. mo. mo. HN.I hof HH.I MM... MH.I mm. MOI mN. m«. m«... .I HmU
me. no... No. mo. 5M... MN... 0M... . MN... MH... N«.I MN. m«. MM... om... Nmm
No.l «o... HN. Ho. MN. MH... MN.I ««.I «H. HH... mo. Mb. M0.I MH. Ham
mo... «0. NH... 0N. Ho. MN... MN... mo... MH. N«... Hm. mm. «H. MN... NMOU
mo. «0. mo... mo. 5N. mof MN.I MN... MM. MN. m«. mm. «H. Mo.l HNOH
Ho. mo. mo. MN. MH.I o«. 5N... 0M. 0M. mo... mm... NM. «N. m«... Nmm
Ho.l NH... mo.l hof OH. MN. H«.I mm. MH. «N. «0.! mm. 00. HN.I Hmm
h m m « M N H h m m « M N H

 

Humane, 3538.6 3389 mmmscasflm

 

mmucHHm> moccocsomm
so NBH>HuonO can mmmcHsmmMHm mo mmcHocoq Houomm

MH OHQMB

83

Table 17

Correlation Matrices of Playfulness and
Creativity Redundancy Variates

 

 

 

 

 

 

 

Unrotated
Creativity
Playful-
ness 1 2 3 4 5 6 7
1 .216 -.142 .552 -.291 . .155 -.061 —.158
2 .155 .150 .363 ' .508 -.114 .109 -.031
3 -.440 -.027 .248 —.016 .116 -.150 -.030
4 .038 -.355 -.056 -.031 -.390 .050 .005
5 -.065 .244 .200 -.153 -.098 -.059 .185
6 -.061 .176 .056 -.033 .034 .260 -.020
7 -.064 .196 -.015 -.151 -.244 -.106 -.108
Rotated
Creativity
Playful-
ness 1 2 3 4 5 6 7
l .755 .000 .000 .000 .000 .000 .000
2 .000 .680 .000 .000 .000 .000 .000
3 .000 .000 .604 .000 .000 .000 .000
4 .000 .000 .000 .493 .000 .000 .000
5 .000 .000 .000 .000 .460 .000 .000
6 .000 .000 .000 .000 .000 -.276 .000
7 .000 .000 .000 .000 .000 .000 .232

 

84

obtained by canonical correlation, make the case for a
relationship between playfulness and creativity much
stronger. The correlations between the first rotated
redundancy variates of playfulness and creativity is
.756.

To test Hypothesis 6, that the relationship
between playfulness and creativity would disappear
when mental age was partialled, the correlations be-
tween playfulness and creativity were recalculated,
partialling mental age. The results which are presented
in Table 18 show little difference with those pre-
viously obtained (compare with Table 15).

Redundancy analysis was performed as before
and the results are presented in Tables 19 and 20.
From Table 19 it can be calculated that the redundancy
variates of playfulness account for 32.3% of the
variance of creativity, while the redundancy variates
of creativity account for 25.5% of the variance of
playfulness. The correlation between the first redun-
dancy variates of playfulness and creativity is .757.
As all these results demonstrate there was little
change in the relationship between playfulness and
creativity, when mental age was partialled, discon-

firming Hypothesis 6.

85

 

oo.H ho. Ho.l ho.l MH.! MH. MH. 5N. MM. MM. MN. MH. mo. MH. MN.
oo.H HM. Hh. M0. MN. HN. Ho. No.l MN. 50. MN. MH. MN. NN.
oo.H MM. Mo.l MH. MH. MH.! MH.I MH. MH.I MH. ho.l NN. Mo.l
oo.H OH.I MN. MN. M0. «0.! M0. M0. ON. NN. N«. «N.
oo.H «0.! Ho. MO.I 0N.I mo.| MN.I MM.I hM.I MH. MH.I
oo.H Mm. MH. 5H. MN. «H.I No. Ho.l No.l Mo.
oo.H Mo. 5H. MN. OH.I MO. MO.I Mo.l «o.l

oo.H MM. NN. NM. 0N. MM. HM. Hh.
oo.H H«. «M. 5M. MM. «0. 0M.
oo.H MM. MM. «M. MD. ON.
oo.H MM. 55. MH. MM.
oo.H 5M. HH.
oo.H HH. NM.
oo.H NM.
oo.H

mdqm
UHmO
mem

DHh

mmazw

Hde

Nmu
HMO
Nmm
Hum
NNOU
HNOH
Nmm
Hmm

 

mflflm mafia. xmgm .HHM mwéflm DEG nfifim NMU .mflu Nmm .flmw NHOU .ERHu Nmmn Hmm

 

 

huH>Hummuo mmmcHoMNMHm

ooHHMHuumm om< chcoz NuH>Humouo
can mmocHsmNch mo unnamed: oc0E¢ mcoHuchHHou

MH OHQMB

86

 

 

 

emf ee. mm. Hm. em. d...- e. me. me.- 8.. ea. mm. sf 3.- mass
em. em.- mmf eef me. 8. ee. 8f ma. 8.. ee. em. 3. eef ono
3.. mm. emf Ef Hm. 3f Sf S. 8.. 2f 8.. if «N. emf x3e
mm. em. 2.. Sf 3. mm. me. 8.. Sf 3. me. 3. em. mef 3m
mm. emf mm. me. me. me. emf ee. 8. me. mm. emf 3.. 3f maze
mm. mm. 3.. mm. ee. Sf 2. me. He. ea. .- Sf emf emf use
me. mm. mmf 5. me. emf ee. me.- 8.. me. 2.. me.- emf mmf zoa
Sf me. me.. mm. me. E. em. em. .2..- mm. ee. me. emf me.- «we
Sf me. e. mm. me. 3.. mm. 8.. $.- ee. eNf em. mef me.- Go
me. mef Sf mef mm. mm.. em. _ Hmf mm. mef Sf me. emf eef mm...
8f 8.. em. Sf 3f 3... me. Sf Sf Sf ee. me. 8. ma. Hm
mef me. me.- mmf Sf em.- em. ee.- 2f emf mef me. em. mmf See
me. we. Sf ee.- emf me.- me. emf 2.. mm. mm... en. es. me.- See
8. me. em. Hmf em. ee. em. mm. mmf ma. ee. mm. mm. mef men
8.. mef maf me. He. em. 2. me. me. mm. 3. me. me. 2f Hem
m e m e m m H m e m e m m A
3938.6 89.3.5, magnified

mmumHHm> Nocmossoom so omHHcHunmm mod Hmucoz
nuHs NuH>HumoHO pom mmmcH5mmmHm mo mmcHocoa Houomm

MH OHQMB

87

Table 20

Correlation Matrices of Playfulness
and Creativity Redundancy Variates
with Mental Age Partialled

 

 

 

 

 

 

 

Unrotated

Creativity
Playful-
ness 1 2 3 4 5 6 7
1 -.014 -.027 -.570 .284 .207 -.137 -.123
2 -.042 .208 -.387 -.533 —.026 .096 -.027
3 .536 -.020 -.102 .007 .040 -.114 -.045
4 .083 .352 .195 .008 .346 -.093 -.023
5 .158 .371 -.124 .152 -.137 .037 .137
6 -.018 -.069 .015 .012 -.149 -.260 .028
7 .021 .168 .040 .101 -.227 -.048 -.117

Rotated

Creativity
Playful-
ness 1 2 3 4 5 6 7
1 .757 .000 .000 .000 .000 .000 .000
2 .000 .684 .000 .000 .000 .000 .000
3 .000 .000 -.626 .000 .000 .000 .000
4 .000 .000 .000 -.502 .000 .000 .000
5 .000 .000 .000 .000 .473 .000 .000
6 .000 .000 .000 .000 .000 .288 .000
7 .000 .000 .000 .000 .000 .000 -.185

 

CHAPTER IV

DISCUSSION

The purpose of the current study was to examine
Dansky and Silverman's (1975) associative fluency hypo-
thesis in a systematic manner. The associative fluency
hypothesis, as formulated by Dansky and Silverman (1975),
states that play creates a mental set which enables the
player to associate fluently and thus do well on
creativity tests since they measure associative fluency.
Since creativity has been defined in their studies as
verbal creativity, the current study was undertaken to
examine the effects of play on other aspects of creati-
vity. This required the use of a variety of creativity
tests and thus treatment of their order as an inde-
pendent variables.

It was therefore quite surprising, given the
strict experimental design, that there was a lack of
significant findings in the experimental analysis. One
problem was the lack of power. It had been expected
that order of creativity test presentation would not
affect the scores on these tests. When it was found
that it did, the number of subjects per cell was reduced
from 15 to 5.

88

89

A second problem was the variance. Attempts were
made to keep the directions comparable. Children who
gave a large number of irrelevant responses when taking
the Alternate Uses Test were excluded from the analysis.
Still there was a lack of homogeneity of variance in
many cases, indicating a good deal of individual varia-
tion within the treatment condition.

While the lack of a significant treatment effect
on the Alternate Uses Test was contrary to expectations,
the sex by treatment interaction offered hope of partial
confirmation. Examination of the means revealed that
girls showed the associative fluency effect, while boys
did not. But examination of the correlations within the
play condition between sex and the other variables
revealed no sex difference.

This suggests that the boys and the girls res-
ponded differently to the control condition. But why
did the boys in the control condition perform better
on the Alternate Uses Test than the girls in that condi-
tion? A number of hypotheses were generated, but none
were satisfactory. For every reasonable explanation
why the boys performed better, there was a reasonable
counter-explanation which would predict better perform-
ance by the girls. It then seems reasonable to think

that this apparent sex difference occurred by chance.

90

What then is the state of the associative
fluency hypothesis? It is probable that the effect
exists, given its finding in three studies (Dansky &
Silverman, 1973, 1975; Sutton-Smith, 1967) and partial
confirmation in the current study. The question
remains why have other studies supported this hypothesis,
while the current study is equivocal on it.

Perhaps the lack of significant findings in the
current study is due to differences between it and those
of Dansky and Silverman. Geographical location is an
unlikely explanation since both studies were done in
, Midwestern cities, Lansing and Toledo. Likewise, it is
éfifficultto imagine a historical change between 1975 and
1978 that would explain the difference.

The age difference in the samples between the
current study and Dansky and Silverman's studies at
first appears to be the explanation. Dansky and Silver-
man's samples were pre-school children who on average
were a year younger than those in the current study.
Torrance (1965) has found that there is a developmental
decline in creativity about the time children begin
kindergarten. This information coupled with the fact
that children in Dansky and Silverman's (1973, 1975)
studies averaged over 10 alternate uses per object,
while those in the current study averaged less than 4

uses, suggests that this decline suppressed the

91

associative fluency effect. But Sutton-Smith (1967)
tested kindergarten-aged children and found the effect,
which negates that suggestion.

It could be argued that the complexity of the
current study wiped out the effect. Two points can be
brought in rebuttal to this argument. First, an
examination of those children who were given the Alter-
nate Uses Test first failed to reveal the hypothesized
effect. Second, if one takes this argument at its
word, it implies that the associative fluency effect
is a weak one.

The weakness of the associative fluency effect
seems to be the most reasonable explanation for the
failure to replicate. The only other remaining explana-
tion is that there was a change in procedure. But for
the most part the procedure in the current study dupli-
cated Dansky and Silverman's, even including the
instructions given the children in the treatment condi-
tions. If there was a difference in procedure, it must
have been slight. This again would attest to the weak-
ness of the effect.

There were other interesting findings in the
current study. The treatment by order interaction was
not hypothesized, but there is a reasonable post-hoc
interpretation. In the current study the creativity

tests were presented to the children not as tests, but

92

as games, following Wallach and Kogan's (1965) sugges-
tixan. Perhaps these tests so presented were in effect
periods of play before the Alternate Uses Test was
presented. If the Barron-Welsh Art Scale is treated
as 5 minutes of play and the Torrance Figure Test as 10
minutes of play, we see the patterns displayed in Figures
8 and 9.

As can be seen in Figures 8 and 9 there was an
increase in the number and uniqueness of uses as the
time spent playing increased up to 20 minutes. There was
a sharp decline in the number and uniqueness for those
who played for 25 minutes. These children could have
become fatigued or bored with the tasks in the experi-
ment. These findings are consistent with Dansky and
Silverman's (1975) associative fluency hypothesis, but
they also suggest that too much play can be detrimental
to performance on creativity tests.

There still remains a major difficulty in
Dansky and Silverman's associative fluency hypothesis.
Why were the near significant effects seen only for
the Alternate Uses Test? It is reasonable that a short
period of play would not affect a personality trait,
such as liking for complexity. But why did play,
essentially a nonverbal activity, at least in the current
study, affect weakly verbal fluency but not affect at

all nonverbal fluency? It would be advisable to perform

93

 

 

 

. ‘ Imitation
- Control

E23 Play
\\\\

l8"

  
  

|6*-

 

WWW

       

WWW

 
 

 

 

 

/'

 

l4

l2
l0 -

8

6

4

2

Uses

25 min

20min

ISmin
FigureB Mean Performance on the AUN given time spent playing

5 min IO min

0 mm

T-rne

94

0:303 2on «E: 5%.. ::< 2: co monogram. cues. meant

 

 

 

   

 

 

 

 

 

 

 

 

£8 mm 58 ON 55m. SE 0. 58 m SE 0 ESP
9 / ...:

1 N

I- i.
.I. . e

-e

.6.

L

L N.

.6.

.2200 I -0.

5:22.. a

.35 ..U L e.

 

95

a study to examine more closely what effect, if any,
play has on nonverbal cognitive skills.

One other finding from the experimental analysis
is of interest. frhe significant treatment effect on
the flexibility score from the Torrance Figural Test
was not predicted, but there is a post-hoc interpreta-
tion for this. The superior performance of children in
the imitation condition on flexibility suggests that it
was flexibility that was modeled in this condition.

The imitation treatment consisted of taking 8 objects and
putting them together in four different ways. This was
certainly more flexibility than most of the children in
the play condition demonstrated in their play.

In summary, the experimental analysis suggests
that the associative fluency effect exists, but it is
a weak effect, disrupted when changes in procedure or
in the complexity of experimentation are made. It would
be worthwhile to examine in a further study the limits
on the associative fluency effect.

There also remains the question why play affects
verbal fluency and not nonverbal fluency. Since non-
verbal fluency has been found to be independent of verbal
fluency, it would follow statistically that play would
not affect nonverbal fluency. But the logic of the
argument is not convincing, since play was primarily

nonverbal. No easy answer comes to mind. Perhaps

96

future studies should encourage verbal response in
order to understand the parameters of play that affect
verbal fluency. From there it would be possible to
suggest conditions under which play would aid nonverbal
fluency.

The second part of the current study was
exploratory in nature. If Dansky and Silverman's (1975)
associative fluency hypothesis is accepted, there still
remains the question what characteristics of play contri-
bute to the mental set which enables the player to asso-
ciate fluently. The findings from the correlation
analysis of children in the play condition suggested
that the imaginativeness and, to some extent, integration,
may be those characteristics.

These two characteristics seem appropriate
because they parallel what occurs during creativity,
when it is defined as an associative process (Mednick,
1962). Creativity, by this definition, involves the
combination of previously distinct elements (i.e.,
integration) in a way to produce a novel result (i.e.,
imaginativeness).

The problem is that the results from the corre-
lational analysis barely reached significance. A weak
relationship is to be expected, because activities

during play are not directly translatable into cognition.

97

(If they were then play would become strictly utili-
tarian, and thus cease to be play.) It would be
advisable to determine if the relationship is weak but
significant by testing a larger sample of children in
the play condition.

As part of the exploratory aspect of the current
study, an attempt was made to build a model for the
relationship between play and creativity. It had been
hoped that a multidimensional scaling solution would
reveal that certain aspects of play are associated with
certain types of creativity. Instead aspects of play
clustered together, as did theetypes of creativity. Upon
afterthought, this was reasonable when one considers
that all aspects of play share a common purposelessness.
Any relation to any type of creativity would be bound
to be tenuous (Sutton-Smith, 1967).

It was still possible to propose a model of the
relationship between play and creativity. Lieberman
(1977) has suggested that playfulness, a personality
trait, is manifested in the way each child plays. This
play can be integrated for use in creativity. Based on
these ideas a path analytic model was proposed, which
fit the data reasonably well.

It can be argued that, since the judges' ratings
of playfulness occurred after their observations of each

child's play, the play caused the playfulness. Whether

98

one model or the other is true cannot be determined within
confines of the current study. I prefer the model
proposed in the path analysis, since it more closely
matches Lieberman's (1977) thoughts about the relation-
ship between playfulness, play, and creativity. In any
event, this model is proposed for experimental testing.

There were a number of interesting findings in
the path analysis in Figure 7. There were two aspects
of playfulness: playfulness-fun and playfulness—
intelligence. While others (Singer, 1966; Sutton-Smith,
1967) have noted that play has two aspects--a cognitive
or applied aspect, and an affective or fun aspect--
Lieberman (1977) has only alluded to it. This was
probably due to her finding that all playfulness items
loaded strongly on one factor. But one advantage of
cluster analysis is that it does not require clusters
to be independent of each other, as factor analysis
does. When one notes that playfulness-fun and playfulness-
intelligence correlated .61 with each other, one can
understand why cluster analysis found two factors,
while factor analysis found only one factor.

The second finding of interest is the corres-
pondence between the playfulness and play clusters. One
of the criticisms made earlier about the Playfulness
Scale is its questionable validity. While the current

study does not establish the validity of the Playfulness

99

Scale, it does demonstrate that the Playfulness Scale
fairly accurately reflects important individual differ~
ences in play, complexity and shifts.

The third finding of interest was that
playfulness-fun and playfulness-intelligence had
different impacts on the complexity and shifts during
play. This finding lends further support to the dis-
tinction between the two aspects of playfulness. In
addition this finding points to two very different
approaches to children's play. Some children appeared
primarily to view play as an intellectual activity intflfieh
they made frequent changes and strove toward increasing
complexity. Other children apparently viewed play
primarily as an activity to be enjoyed. They became
easily satisfied playing with a small set of objects,
with which they make very few changes. This is not an
all-or—none dichotomy;children showed some degree of
each as the high correlation between the two aspects of
playfulness demonstrates.

The redundancy analysis demonstrated a degree
of commonality between playfulness and creativity, con-
sistent with other studies.(Durrett & Huffman, 1968;
Lieberman, 1965). This does not contradict the results
from the path analysis. As can be seen in Table 14, the
path model is weakest in predicting the relationship

between playfulness and creativity.

100

Contrary to Singer and Rummo (1973), this
commonality remained constant when mental age was
partialled. Perhaps this difference is due to some
difference in procedure between the current study and
Lieberman's (1965). The most obvious difference is the
familiarity of the judges with the children being rated.
In the current study the judges saw the children for only
10 minutes before making their judgments; in Lieberman's
(1965) study the judges were the children's teachers and
knew them quite well. It seems possible that the results
found by Singer and Rummo (1973) were due to a halo
effect. Children who are known by their teachers to be
creative and intelligence would also be assumed to be
playful. The observers in the current study did not
know the children well enough to be affected in such a
way.

In summary, the current study makes two major
points about playfulness. First, playfulness appears to
have some validity in describing play activities.
Second, there are two distinct aspects of playfulness-—
a cognitive or applied aspect, and an affective or fun
aspect.

With regard to play, the path analysis demon-
strates two related but distinct aspects of play--
complexity and shifts. Both aspects have been noted

before by others. Bishop and Chace (1971) in examining

101

the creativity in play paid close attention to
complexity of the designs children made in playing with
geometric shapes. Kaggan et al's (1964) notion of
impulsivity is virtually synonymous with the current
study's use of shifts.

What is intriguing here is that certain objects
lent themselves more closely to one style or the other.
Matchboxes and 3 x 5 cards were associated with shifts,
while paper clips and spools were associated with com-
plexity. Researchers on style of play must be aware
that their choice of toys may influence the style of
play children show.

Besides pointing out the different styles of
play, the current study reemphasized a point made by
Sutton-Smith (1967) that the relationship between play
and cognitive development is tenuous, because of the
purposeless nature of play. Any study of play and
cognition will have to be content with low correlations.
This means that a large sample of children will be
required to determine if the low correlations are sig-
nificantly different from zero.

Researchers of play should not be discouraged
by the weak relationship between play and cognition.
Rather they should examine aspects of play, such as
complexity and shifts, that may be related to cognition.

Care must be exercised that in trying to strengthen the

102

relationship between play and cognition, such as by
making the play situation similar to the skill being
learned, the playful character of play is not lost.

In short, play is weakly related to cognition.
This weak relationship should not dishearten researchers.
But rather reserachers should use their ingenuity to
look at specific aspects of play that may be related to
cognition.

With respect to creativity, there are two choices
of definition. One choice is to accept the definition
of creativity as verbal associative fluency, as Dansky
and Silverman (1975) have, and thus have a one-to-one
relationship between play and creativity. The other is
to define creativity, as it is defined in the current
study, as an associative process which includes verbal
and nonverbal cognitive skills, as well as personality
qualities, and thus is weakly related to play.

The second choice is preferred for a number of
reasons. First, it appears more realistic in terms of
the relationship between play and cognitive development,
as discussed by Sutton-Smith (1967). Second, our
naive notions of creative persons include both painters
and writers. Creativity is a multi-faceted ability
(Anastasi & Schaefer, 1971). Finally, the cluster

analysis demonstrated that verbaland.nonverbal creativity

103

are more similar to each other than they are to play
or playfulness.

The definition used in the current study also
attempts to reconcile two views of creativity within
psychology. To personality theorists creativity is a
trait which distinguishes one person from another. To
cognitive theorists creativity is a skill that all
persons have. The point of the definition in the current
study is that the two views are interrelated. While
the current study views the personality dimension as
resulting from the cognitive dimension, others could view
it the other way around. No matter which viewpoint is
believed, personality and cognitive theorists should
include the other viewpoint within their theory. Ex-
perimental study could help answer which results from
which.

In conclusion then, creativity, defined as an
associative process that involves verbal and nonverbal
skills that includes personality traits, is preferred
for two important reasons. First, it more adequately
reflects the hypothesized tenuous relationship between
play and cognitive development than does the restricted
definition used by Dansky and Silverman (1975). Second,
it attempts to reconcile viewpoints within psychology
that conceive of creativity as either a personality

trait or a cognitive skill.

104

Finally, what are the next steps in studying the
relationship between play and creativity? A number of
variations of the current study have been or can be
suggested, including closer examination of children in
the play condition by having them verbalize, varying the
content of the pictures used for coloring in the control
condition to determine why boys in that condition were
more creative than those in the play and imitation
condition, and careful study of children at play for
long periods of time to determine why there was such a
sudden drOp in creativity after 25 minutes of play.

More generally it appears appropriate that
future studies of play and creativity return to a more
naturalistic approach such as that used by Bishop and
Chace (1971) and J. E. Johnson (1976). Such an approach
could involve careful observation of the child during
play at home or in school and administration of crea-
tivity tests that more adequately reflect that environ-
ment, e.g., alternate uses for objects in that
environment, simple or complex drawings that children
of that age might produce. With this type of approach
using a large sample of children, the playful character
of play and creativity might be preserved, while further

exploring the relationship between the two.

REFERENCE NOTES

105

REFERENCE NOTES

1. Lazar, B. S. Creativity and regression on the
Rorschach. Paper presented at the 85th Annual
Convention of the American Psychological
Association Convention, San Francisco, 1977.

2. Hunter, J. E., & Cohen, S. H. PACKAGE: A FORTRAN
IV program. E. Lansing, Michigan: Michigan
State University, Computer Institute for
Social Science Research, 1971.

3. Thissen, M., & Van den WOllenberg, A. L. REDANAL:
A FORTRAN IV G/H program for redundancy analysis
(Research Bulletin 26). Nijmegen, the
Netherlands: University of Nijmegen, Depart-
ment of Mathematical Psychology, 1975.

106

REFERENCES

107

REFERENCES

Anastasi, A., & Schaefer, C. E. Biographical correlates
of artistic and literary creativity in adoles-
cent girls. Journal of Applied Psychology,
1969, 53, 267-273.

 

Anastasi, A., & Schaefer, C. E. Notes on the concept of
creativity and intelligence. Journal of
Creative Behavior, 1971, 5, 113-116.

 

 

Barron, F., & Welsh, G. S. Barron-Welsh Art Scale.

Palo Alto, Ca.: Consulting PsyChologists Press,
1963.

 

Bellak, L. Creativity: Some random notes for a
systematic consideration. Journal of Projective
Techniques, 1958, 32,1363-380.

 

 

Bishop, D. W., & Chace, C. A. Parental conceptual
systems, home play environment, and potential
creativity in children. Journal of Experimental

 

Child Psychology, 1971, P.- 318—338.

 

Burt, C. The psychology of creative ability. British
Journal of Educational Psychology, 1962, 22,
292-298.

 

Campbell, D. T., & Fiske, D. W. Convergent and dis-
criminant validation by the multitrait-
multimethod matrix. Psychological Bulletin,
1959, 56, 81-105.

 

Dansky, J. L., & Silverman, I. W. Effects of play on
associative fluency in pre-school aged children.
Developmental Psychology, 1973, 2, 38-43.

 

Dansky, J. L., & Silverman, I. W. Play: A general
facilitator of associative fluency. Develop-
mental Psychology, 1975, 11, 104.

 

Durrett, E., & Huffman, W. Playfulness and divergent
thinking among Mexican-American children.
Journal of Home Economics, 1968, 62,
355—358.

 

108

109

Feitelson, D., & Ross, G. S. The neglected factor--
play. Human Development, 1973, 16, 202-223.

 

Finn, J. D. A general model for multivariate analysis.
New York: Holt, Rinehart, & Winston, 1974.

Freud, S. [The relation of the poet to day-dreaming.)
In J. Riviere (Ed. and trans.), Collected
a ers (Vol. 4). New York: Basic Books,

 

Freyberg, J. T. Increasing the imaginative play of
urban disadvantaged kindergarten children
through systematic training. In J. L. Singer,
The child's world of make-believe. New York:
Academic Press, 1973.

 

Garvey, C. Play. Cambridge, Mass.: Harvard University
Press, 1977.

Getzels, J. W., & Jackson, P. W. Creativity and
intelligence. New York: Wiley, 1962.

 

 

Groos, K. The play of animals. New York: D. Appleton,
1898.

 

Guilford, J. P. The structure of intellect. Psycho-
logical Bulletin, 1956, 5}, 267-293.

 

Guilford, J. P. Creative abilities in the arts.
Psychological Review, 1957, 64, 110-118.

 

Guilford, J. P. Some misconceptions regarding measure-
ment of creative talents. Journal of Creative
Behavior, 1971,§, 77-87.

Hutt, C., & Bhavani, R. Predictions from play. Nature,
1972, 237, 171-172.

Johnson, J. E. Relations of divergent thinking and
intelligence test scores with social and non-
social make-believe play of preschool children.
Child Development, 1976, 41, 1200-1203.

 

Johnson, 8. C. Hierarchical clustering schemas.
Psychometrika, 1967, 22, 241-254.

 

110

Kagan, J., Rosman, B. L., Day, D., Albert, J., &
Phillips, W. Information processing in the
child: Significance of analytic and reflective
attitudes. Psychological Monegraphs, 1964, 18,
(1, Whole No. 578).

Klinger, E. Development of imaginative behavior:
Implications of play for a theory of fantasy.
Psychological Bulletin, 1969, 12, 277-298.

Kris, E. Psychoanalytic exploratione in art. New York:
International Universities Press, 1952.

Kruskal, J. B. Multidimensional scaling by Optimizing
goodness of fit to a nonmetric hypothesis.

Lieberman, J. N. Playfulness and divercrent‘thinking: An
investigation of their relationship at the
kindergarten level. (Doctoral dissertation,
Columbia University, 1964). Dissertation
Abstracts, 1964, 25, 3677, (University Micro-
films No. 64-9891YT

 

Lieberman, J. N. The relationship between playfulness
and divergent thinking at the kindergarten
level. Journal of Genetic Psycholoqv- 1965,
.101, 219-224T0

Lieberman, J. N. Personality traits in adolescence: An
investigation of playfulness-nonplayfulness in
the high-school setting. NIMH Terminal Report,
December, 1967. (ERIC Document Reproduction
Service No. ED 032 584.)

Lieberman, J. N. Playfulness: Its relationship to
imagination and creatiVity. New York:
Academic Press, 1977.

Maccoby, E. E., & Jacklin, C. N. The psychology of sex
differences. Stanford, Ca.: Stanford UniVer-

sity Press, 1974.

 

Manosevitz, M., Fling, S., & Prentice, N. M. Imaginary
companions in young children: Relationships
with intelligence, creativity and waiting abil-
ity. Journal of Child Psychology and Psychiatry,

1977, Igj‘75378.

lll

Mednick, S. A. The associative basis of the creative
process. Psychological Review, 1962, 69, 220-
232. ‘—

Piaget, J. Play, dreams and imitation in childhood.
New York: W. W. Norton, 1962.

 

Piaget, J. The moral judgment of the child. New York:
Free Press, 1965.

 

Piaget, J. Response to Brian Sutton-Smith. Psycho-
logical Review, 1966, 22, 111-112.

 

Rosen, C. E. The effects of sociodramatic play on
problem-solving behavior among culturally

disadvantaged children. Child Development,
1974, 15, 920-927.

 

Saltz, E., & Johnson, J. Training for thematic-fantasy
play in culturally disadvantaged children:
Preliminary results. Journal of Educational
Psychology, 1974, fig, 623-630.

 

 

Schaefer, C. E. Imaginary cOmpanions and creative
adolescents. Developmental Psychology, 1969,
1, 747-749.

 

Schaefer, C. E., & Anastasi, A. A biographical inventory
for identifying creativity in adolescent boys.
Journal of Applied Psychology, 1968, 52, 42-48.

 

Singer, D. L., & Rummo, J. Ideational creativity and
behavioral style in kindergarten age children.
Developmental Psychology, 1973, 8, 154-161.

 

Singer, D. L., & Whiton, M. B. Ideational creativity and
expressive aspects of human figure drawing in
kindergarten age children. Developmental
Psychology, 1971, 4, 366-369.

 

 

Singer, J. L. Daydreaming: An introduction to the
experimental study of inner experience. New
York: Random House, 1966.

 

 

Singer, J. L. The child's world of make-believe.
New York: Academic Press, 1973.

 

112

Singer, J. L., & Antrobus, J. S. A factor-analytic
study of daydreaming and conceptually-related
cognitive and personality variables. Perceptual
and Motor Skills, 1963, 21, 187-209.

 

 

Slobin, D. I. The fruits of the first season: A dis-
cussion of the role of play in childhood.
Journal of Humanistic Psychology: 1964, 2, 59-
79.

 

Smilansky, S. The effects of sociodramtic play on dis-
advantaged preschool children. New York:
Wiley, 1968.

 

 

Spencer, H. Principles of psycholo (3rd Ed., Vol. 2).
New York: D. Appleton, 189 .

 

Starkweather, E. K. Problems in the measurement of
creativity in preschool children. Journal of
Educational Measurement , 1964, 2, 109-113.

 

 

Starkweather, E. K. Creativity research instruments
designed for use with preschool children.
Journal of Creative Behavior, 1971, 2, 245-255.

 

Sutton-Smith, B. Piaget on play: A critique. Psycho-
logical Review, 1966, 12, 104-110.

 

Sutton-Smith, B. The role of play in cognitive develop-
ment. Young Children, 1967, 22, 361-370.

 

Sutton-Smith, B. A reply to Piaget: A play theory of
copy. In R. E. Herron & B. Sutton-Smith (Eds.).
Child's play. New York: Wiley, 1971.

 

Sylva, K., Bruner, J. S., & Genova, P. The role of play
in problem-solving of children 3-5 years old.
In J. S. Bruner, A. Jolly, & K. Sylva (Eds.),
Play: Its role in development and evolution.
Hammondsworth, England: Penguin, 1976.

 

Thorndike, R. L. The measurement of creativity.
Teachers College Record, 1963, fig, 422-424.

 

Torrance, E. P. Priming creative thinking in the primary
grades. Elementary School Journal, 1961, eg,
34-41.

 

Torrance, E. P. Rewarding creative behavior. Englewood

 

113

Torrance, E. P. Torrance tests of creative thinking:
Directions manual and scoring_guide; vefbal
test, booklet A, research edition. Princeton,
N.J.: Personnel Press, 1966 (a).

Torrance, E. P. Torrance tests of creative thinking:
Directions manual and scoring guide; figural
test, booklet A, research edition. Princeton,
N.J.: Personnel Press, 1966 (bl:

Van den Wollenberg, A. L. Redundancy analysis: An
alternative for canonical correlation analysis.
Psychometrika, 1977, 22, 207-219.

 

Vygotsky, L. S. Play and its role in the mental
development of the child. Soviet Psychology,
1967, 2 (3), 6-18.

Wallach, M. A. Creativity. In P. H. Mussen (Ed.),
Carmichael's manual of child psychology (3rd
Ed., Vol. 1). New York: Wiley, 1970.

Wallach, M. A., & Kogan, N. Modes of thinking in young
children. New York:‘ Holt, Rinehart, & Winston,

1965.

Ward, W. C. Creativity in young children. Child Develop-

Ward, W. C. Creativity and environmental cues in nursery
school children. Developmental Psychology,
1969, 2, 543-547.

Weisler, A., & McCall, R. B. Exploration and play:
Résumé and redirection. American Psychologist,
1976, 22, 492-508.

Weissman, P. Ego functions in creativity. Ps cho-
therapy and Psychosomatics, 1967, 22, 273-285.

APPENDIX A

CORRELATIONS AMONG MEASURES
OF CREATIVITY, PLAYFUL-

NESS, AND PLAY

114

115

IPPIIDIX A

Cottalatiaaa Alana Iaaauraa at Craativtty, Playfulnaaa, and Play

 

 

 

 

III II .1 U1 '2 U2 "5 05 '5 05 An. ADD 0.00 PL” 'le 0'10 ILA, P81 P52 0021 0072 831 532
in .u ,
.1 -.25 .05
m -.., .05 0"
.1 -.15 91° a.2 a1
02 -.15 .05 .75 .75 .55
'5 .05 C 01 .55 .50 .50 .55
05 .02 00 .25 .25 .55 .20 .55
.5 -.05 .25 .75 .75 .55 .50 .50 .50
05 '.05 .25 .72 .77 .55 .55 .55 .57 .55
m -al, .10 a’J 0,0 a’l ... a“ .52 a.‘ a.)
M --17 al‘ a.’ ’3 a.’ a.‘ .50 .55 0.5 a.‘ 9"
m -.00 .00 .05 -.01 .10 all -011 -.15 -all -a1‘ .02 -1“
PL” .17 .00 .15 .15 .15 .15 .15 .15 .25 .51 .20 .25 -.11
m '.25 a2: .10 .15 .12 .07 al‘ 0" a21 .50 015 .1, -aos a‘l
0.55 .25 -.05 .11 .12 .15 .11 .25 .27 .25 .51 .20 . .05 .71 .55
M '1‘ .05 -010 -.00 02‘ .21 a2. .20 .25 a). a1‘ ‘1‘ -.17 -.0. -.01 a0,
'81 a.) a2) -010 .01 -.15 002 .10 a“ a2: .22 -003 .00 -.12 a2: -.01 .20 .2.
P32 a“ .3, -006 -003 -.02 a0; -a0‘ .01 I12 .15 -a°2 0°) .20 a.° a2. 02‘ a1, .‘5
JOYI .05 .22 .01 .05 -.15 -.17 .05 .05 .15 .00 -.02 .02 0.55 .21 -.05 .12 .10 . .17
.7022 “.07 .23 '.05 0°, '.05 --1. '20 a1, .25 '1, .05 .05 -03‘ a5, a3, .21 017 a). a1. 0"
$31 .05 .00 '.00 ‘.05 -al‘ -.21 -010 .51, 0°: -00, “.10 -.25 ..25 .05 -al. .07 .25 a,7 .17 a7, a‘:
832 -.15 .22 .22 .25 .25 .15 .25 .15 .57 .50 .25 .27 -.05 .05 .25 .25 .55 .25 .15 .57 .50 .55
C51 - 03 .25 .01 .21 .1. .10 .50 013 .23 a21 a1, .50 -'1. ..05 -90. -.05 ’3‘ .3. .12 a‘: .55 .55 a“
C32 .55 .55 .02 .11 .05 .1 .22 .05 .25 .25 .12 .15 -.02 .05 -.02 -.02 .27 .72 .55 .55 .20 .52 .25
IN? .15 .55 .00 .07 .15 .12 .52 .55 .25 .25 .15 .15 .00 -.10 -.05 -.05 .55 .55 .55 .55 .55 .55 .55
PA -.15 .51 .15 .15 .05 .05 .10 .25 .17 .15 .15 .15 -.15 .01 .05 -.05 ..05 -.20 -.15 .05 -.02 .05 -.05
PC -.15 .27 .15 .15 -.05 .01 -.22 -.25 .15 .05 .05 .05 .15 -.20 -.15 -.55 -.05 .25 .25 .55 .11 .55 .15
cm 01. .22 -007 “.00 -als -00, -al, -010 .05 a1‘ -010 -.°5 al‘ -.05 -01. -001 ‘00, a‘l a o ..01 -01 -a1’ -.20
I! .21 .07 .25 .21 .22 .55 .05 .05 .51 .55 .27 .52 .07 .07 -.15 .05 .17 .55 .51 07 .01 -.02 .07
Cu .12 a0, - 12 -010 - 03 0°, col .10 .05 .05 -005 -001 ..3‘ -g“ -.00 -.05 -003 .32 .05 .21 .01 to, -al,
" .25 .55 00 no, -.20 -all -a0. .03 0°. .1, -.05 0°: .05 a1: .2, 011 ..21 .33 .55 a1, .05 .05 .10
CL -.05 a1, 1‘ .02 .00 00’ .a3‘ ‘an -all “.05 -003 -a°‘ a3, a .11 -a°2 -a22 -a3‘ ~15 - J. -.55 -al‘ -a16
CP -.11 -.05 0.02 -.05 .12 .05 .00 -.02 0.05 -.00 .02 -.02 .55 -.27 -.20 -.22 .22 -.52 -.05 -.12 .02 .05 .00
'L -al’ -.50 . 05 '.01 -.07 -a11 00‘ .07 .13 .0, -a0: -.01 -017 '.05 '.05 .01 -.10 '11 -.05 -013 0” .a06 .12
m .12 01° .05 --03 .00 .1: -a23 ”a2. a1: .1: .00 .01 03‘ to. -.05 00‘ -.01 .11 v‘J -011 -.12 .10 no)
‘3’ 'al‘ -a03 - 03 00‘ .03 -Io’ 02 .21 -00, -507 .05 to: -0” -001 no, -aO, 0°, -.15 -.50 all al‘ ’00, ‘.01
m n13 - 51 -020 -al‘ -aJJ -031 -017 -.15 -.25 -03. -.25 -a2. -.20 .13 -al‘ .01 .021 -.02 -522 a0, -01. -.05 -.51
m -.15 .10 - 0‘ -a21 -00, '.02 -a1. -.05 -all -012 -011 -a5‘ .20 ‘00. all -a°7 'a’O -.15 .05 -.2. -al‘ -015 .00
runs 07 .21 .15 .55 .52 .25 .55 .15 .52 .57 .31 .55 -.15 .01 -.02 .05 .57 .15 .05 .25 .25 .22 .15
m .05 ~23 all .05 '.01 all -02. -al’ 01‘ al‘ a0: .05 a3) -... -ao’ -00. -907 .25 a‘, -.05 -a13 a0, .01
0881" 017 01‘ - 03 a“ -017 -a°. -.., -.15 .1. at: -00‘ .00 ".05 .02 a0) -a°7 ‘50. .55 a3. .30 all .11 .00
m1" .1: .1. -003 .00 -al. -a07 “.05 ’0‘. a15 a1. -.05 1°, -.05 .10 .03 -.05 -a“ a‘3 .27 0‘1 al‘ .1. .00

. a)?

C31 C32 2'! PI PC (INI7 I. ‘3': IP Cl CP Pl IIHIIFIIP IIP 5" 5"0 25:2 CIIIPT
C52 .55
2.? .57 .55
PA 0.05 .01 .20
PC .12 .55 .55 .50
C519 -.07 .22 .05 -.20 .12
I. .05 .55 .05 -.25 .25 .55
CI! .07 .17 .15 .07 .05 .21 .55
” -l13 a.) .55 03° 0" p22 .00 '.°5
CL -I‘. -020 -al‘ '22 .25 ‘.02 -02‘ -..; 0.,
CP al‘ -.05 a2? -.05 .15 -022 '.50 -051 -00. .13
n -00. -017 -.22 -a‘l -alo .20 I“ -00. -.12 ‘ajl a
m ’02) 01: 01‘ -013 a‘: a1, a" -00. a“ .50 .35 .1.
m .25 -aa‘ -411 a1: -0‘1 -023 .03) 01° '.52 -a‘7 -.J, -.20 ...,
w -01, ‘.25 -n“ '.15 -a17 -al‘ -01, .00. -a2° -.20 -002 .2, .922 .10
m -.55 -.25 -013 '.05 .12 51. '11 01. .20 .3. -a13 0° 0 -031 -05:
1.55 70 .50 .55 .15 -.02 -.12 -.07 -.17 -.10 -.25 .17 -.15 -.25 .50 -.07 -.51
w-a1’ 027 '2‘ a0. .61 .5. a‘l I02 0‘ .50 a” .” 0’: n.” -9” a” O.“
m to, 0“ I1. -021 0‘0 '51 '3’ '21 .55 ’oo -0” a a” -0” -.01 002 -.01 a“
m .21 .55 .20 -.27 .52 .50 .51 .15 .51 -.15 -.21 . .32 -.21 .15 ...” ‘1‘ .3 .”

Correlatloaa uhaaa abaoluta valua 1a

gtaata: than or aqua! to .36 ara aigaiticant to tha .05 laval.

nIcumnN snare UNIV. LIBRQRIES
IIHI ”WI I” ’III WIN” ll “W 1| W IIIIH W "WI
31293103438044