KI-SOON HAN

Domain-Specificity of Creativity in
Young Children: How Quantitative
and Qualitative Data Support It

ABSTRACT One of the most controversial issues in contemporary research

of creativity, whether a person’s creativity is domain-specific
or domain-general, was investigated in the present study. This
study is composed of two parts, Study 1 and Study 2. In study
1, the relationships among 109 children’s creative perfor-
mances in three domains, and the relationships between those
children’s general creative thinking skills and their creative
performances in three domains have been examined. Study 2
examines how the domain-specificity and -generality issue is
addressed in individual children via case studies of three highly
creative children, hoping to provide enriching and qualitative
specification to the quantitative data of the present study. In
both studies, children’s performances in language, art, and
math domains were respectively judged by three experts who
rated children’s creativity on story-telling, collage making, and
math word-problem creating tasks. Children’s general creative
thinking skills were assessed by a battery of two divergent think-
ing tests, including the Wallach-Kogan Creativity Test
(Wallach & Kogan, 1965) and the Real World Divergent Think-
ing Test adapted from Okuda, Runco, and Berger (1991). The
findings of this study support the position that creative ability
in young children is rather (but not absolutely) domain-specific.
INTRODUCTION This study rests on one of the most controversial issues in
contemporary research of creativity. Is a person’s creativity
domain-general or domain-specific? The supposition that hu-
man creativity is a generalized ability — similar irrespective of
the kind of discipline or subject matter involved — has guided
much of the research and theory development in the study of
creativity over the last 50 years (Barron, 1988; Cramond, 1994;

117 Volume 37 Number 2 Second Quarter 2003

37-2-03 pages.p65 117 3/20/03, 8:35 AM

 Domain-specificity of creativity in young children

Guilford, 1967; Hocevar, 1980; Milgram & Milgram, 1976;

Plucker, 1998; Runco, 1986; Torrance, 1966, 1988). However,
a paradigm shift toward a domain-specific view of creativity is
occurring in the study of creativity. Recent theoretical
(Csikszentmihalyi, 1988; Feldman, 1980, 1982, 1994; Gardner,
1983) and empirical (Baer, 1991, 1993, 1994a; Gardner, 1983;
Runco, 1989) literature suggests that creativity may be more
a specific trait than was once believed.
Driving the shift from the general view of creativity toward
the specific has been an increasing objection to the heavy
emphasis placed on divergent thinking measures as an index
of a general capacity for creativity (e.g., Anastasi, 1982; Baer,
1994b; Brown, 1989; Milgram, 1990; Runco & Nemiro, 1994).
Although under some conditions, divergent thinking tests may
predict creative achievement, the validity of divergent think-
ing tests has been questioned due to the concern that high
scores on a divergent thinking test may not always predict cre-
ative achievement (e.g, Baer, 1993). There is also limited con-
vincing evidence that children judged creative in one domain/
discipline necessarily display strong divergent thinking skills
(Gardner, 1993a, 1993b). Researchers have started to report
that divergent thinking is only one aspect of creativity, and that
individual knowledge domains do play an important role in
understanding and measuring creativity (Csikszentmihalyi,
1990; Feldhusen & Goh, 1995; Feldman, 1980; Gardner, 1983).
Although the research on ‘domain-specificity’ of creativity
is regarded as an important breakthrough by some research-
ers (e.g., Runco & Nemiro, 1994), the debate over the domain-
specificity and -generality still remain controversial, ill-defined,
and clearly complex owing to the limited number of empirical
studies. Studies also lack a clear and agreed-upon result and
conclusion, and are open to a plethora of individual viewpoints.
Tardif and Sternberg (1988) indicated “it is generally acknowl-
edged that people are creative within particular domains of
endeavor . . . when the issue of domain-specificity occurs . . .
much less agreement ensues (p.433).” Creativity has been
argued as domain-general, as domain-specific, or as both de-
pending upon the restricted theory and/or data base used. For
example, some contemporary researchers of creativity seem
to agree on the domain-specificity of creativity at least theo-
retically (Csikszentmihalyi, 1990; Gardner, 1993a, 1993b);
however, their position lacks sufficient empirical support. There
also have been research studies that introduced creativity as
both domain-general and -specific construct and empirically

118

37-2-03 pages.p65 118 3/20/03, 8:35 AM

 Journal of Creative Behavior

distinguished them (Hong, Milgram & Gorsky, 1995), but their

conclusions lacked solid theoretical foundations and further
research support. In addition, the dichotomy in research
findings may be caused by methodological problems, such as
methods and statistics used. Bivariate analyses appear to fa-
vor specificity while multivariate analyses tend to find evidence
for the domain generality of creativity. Performance-based
assessments often produce evidence of specificity and self-
report scales suggest evidence for generality. As Plucker (1998)
suggested that the idea of domain-specificity suffer from con-
ceptual, methodological, or logical oversights. Further evidence
is essential to inform the debate. Followings are some of the
limitations addressed in previous studies.
First, most studies that have explored the domain issue of
creativity were limited to the use of a self-report scale as a
measure of creative performance in diverse domains resulting
in support for creativity’s domain-generality (Hocevar, 1980;
Milgram & Milgram, 1976; Runco, 1986, 1987). Although some
researchers suggest that self-report scales are conceptually and
psychometrically reasonable measures especially when the
creative accomplishment is measured (Hocevar, 1980), self-
report scales have been criticized for their lack of reliability
and validity in assessing creative abilities (especially in young
children) (Brown, 1989). Performance-based or product-based
assessments using expert consensus (Amabile, 1983, 1996)
have been strongly recommended for assessing children’ cre-
ative abilities in diverse domains. Although performance-based
assessment is not without its problems, such as lack of gener-
ality, increased cost, more involved scoring and so on, perfor-
mance-based assessment is well accepted because it embeds
assessment in meaningful, intelligence-fair, real-world activi-
ties. However, only a few studies to-date have used performance-
based assessments in limited number of domains (e.g., Baer,
1994a; Runco, 1989), and very few studies have used expert
consensus in evaluating children’s performance-based prod-
ucts to explore the question of creativity’s domain specificity–
generality.
Second, although studies have shown interest in looking at
the relationship between divergent thinking and creative per-
formances, few studies have paid attention to the relationships
among children’s diverse creative performances in diverse
domains. Researchers have been mainly interested in criterion-
related validity (concurrent or predictive) of divergent think-
ing measures in predicting creative performances. However,

119

37-2-03 pages.p65 119 3/20/03, 8:35 AM

 Domain-specificity of creativity in young children

exploration of inter-relations among a person’s creative per-

formances in different domains is essential in investigating the
question of domain-specificity and -generality of creativity.
Third, no study to-date has utilized a battery of divergent
thinking tests to explore the domain-specificity and domain-
generality issue. Most studies have relied on a single divergent
thinking test, thereby lacking the support of an adequate data-
base. Since the correlations between various divergent think-
ing tests suggest that each taps various aspects of divergent
thinking (Kogan, 1994), the use of a battery of divergent think-
ing tests would be necessary for careful examination of the
domain issue.
Fourth, it has been suggested that divergent thinking tests
would be more predictive of real-world creative performances
if they contained problems children might encounter in their
school or home settings (Hong & Milgram, 1991; Okuda,
Runco, & Berger, 1991; Runco, 1993; Runco & Okuda, 1988).
However, the studies utilizing real-world divergent thinking tasks
have correlated results with only self-report scales as measures
of creative performances in diverse domains. There has been
no published study that looks at the relationship between a
real-world divergent thinking test and children’s creative per-
formances in diverse domains utilizing performance-based
assessments.
Fifth, most studies in this area have involved older subjects
(e.g., high school students, college students). Published stud-
ies on young children (age 3 to 8 years) in this area have been
meager, and limited to small sample sizes (e.g., Baer, 1991).
Finally, most studies on the domain issue of creativity have
examined the problem only quantitatively, lacking the support
of rich qualitative data of the issue. Little attention has been
paid how the domain issue can be addressed in individual
creative children, although rich qualitative data might provide
enriching and qualitative specification to the quantitative data
because of the complexity and subjectivity of the question
itself.
Thus, the purpose of the present study is to investigate the
issue of domain-specificity and domain-generality of creativ-
ity in young children both in quantitative and qualitative ways.
Study 1 examined (1) the relationships among children’s
creative performances in three domains, and (2) the relation-
ships between children’s general creative thinking skills and
children’s creative performances in three domains. Children’s
performances in language, art, and math domains were respec-

120

37-2-03 pages.p65 120 3/20/03, 8:35 AM

 Journal of Creative Behavior

tively judged by three experts who rated children’s creativity

on story-telling, collage making, and math word-problem cre-
ating tasks. Children’s general creative thinking skills were
assessed by a battery of two divergent thinking tests, includ-
ing the Wallach-Kogan Creativity Test (Wallach & Kogan,
1965) and the Real World Divergent Thinking Test adapted
from Okuda, Runco, and Berger (1991).
Study 2 examines how the domain-specificity and -general-
ity are addressed in individual children via case studies of three
creative children, hoping to provide enriching and qualitative
specification to the quantitative data of the present study. The
results of both studies would contribute to our understanding
of the nature of creativity in young children. The findings might
permit reconsideration of assessment practices in the field of
gifted education. Since the domain-generality and domain-
specificity issues are directly related to identification and edu-
cational practices for creative children, careful examination on
these issues are critical.

STUDY 1 One hundred and nine second grade children from five urban
Subjects elementary schools, 53 (49%) boys and 56 (51%) girls, partici-
pated in the present study. The age range of the subjects var-
ied from 7.01 (85 months) to 8.09 (104 months) years, with a
mean of 7.10 years (93.7 months) and the standard deviation
of 4.4 months.
Instruments Wallach-Kogan Creativity Test. Two verbal subtests (Al-
ternate Uses and Similarities) and one nonverbal subtest
(Pattern Meanings) of the Wallach-Kogan Creativity Test
were selected for the present study. Each verbal or nonverbal
subtests had three items in it. In addition, in the present study,
the Wallach-Kogan Creativity Test was modified to include
problem-finding tasks. It has been suggested that problem-
finding tasks enhance the validity of divergent thinking tests
as measures of creativity (Wakefield, 1985, 1992). Therefore,
three items in each subtest were composed of two problem-
solving tasks and one problem-finding task. The problem-
solving tasks in each subtest asked children to tell all the dif-
ferent ways they could use an object, how two objects are alike,
or all the things a pattern could be. The problem-finding task
in each subtest asked children to generate a problem and then
provide a solution to it. For example, a problem-finding task in
the Pattern Meanings subtest was like “Here is a blank card
and a pencil. Make a pattern of your own, then tell me all the
different things it could be.”

121

37-2-03 pages.p65 121 3/20/03, 8:35 AM

 Domain-specificity of creativity in young children

Scoring followed the standard scheme suggested in the test

manual. Fluency (number of responses) and originality (num-
ber of unique responses) scores for each problem-solving
task, and a fluency score for each problem-finding task were
generated. The scores were summed across the subtests to
yield scores for Wallach-Kogan problem-solving fluency
(WKF), problem-solving originality (WKO), and problem-
finding fluency (WKPF).
Real-World Divergent Thinking Test. A Real-World Diver-
gent Thinking Test, adapted from Okuda et al. (1991), was
used in this study. The adaptation for the present study con-
ceded situations and problems relevant for second grade chil-
dren. In this study, four real-world divergent thinking tasks (two
problem-solving and two problem-finding tasks) were used. All
tasks were related to a school situation. In the real-world prob-
lem solving tasks, students heard problems related to school
life, read aloud by the examiner. Students were asked to pro-
vide as many solutions as possible. For the real-world prob-
lem-finding tasks, students heard problematic vignettes about
school life, and were asked to list all the problems they can
think of in each of these settings. The problem-solving tasks
were scored for fluency (number of responses given, RWF)
and originality (number of unique responses produced by less
than 5% of children in the sample, RWO), and the problem-
finding tasks were scored only for fluency (RWPF). Scoring
for the Wallach-Kogan Creativity Test and the Real-World
Divergent Thinking Test was done by the first author and two
trained graduate students. The agreement between the three
scorers was 95% or higher across all subtests of the two diver-
gent thinking tests.
Performance-based assessments. Three performance-
based assessments were utilized in this study. The three
assessments included: a story-telling task (language), a col-
lage-making task (art), and a math word-problem task (math).
All tasks were selected from various sources using Amabile’s
(1983, 1996) guidelines for selecting appropriate tasks for a
consensual assessment. Amabile (1996) indicated that an
appropriate task for a consensual assessment should meet the
following requirements: (1) the task must lead to some prod-
uct that can be available to appropriate judges for evaluation,
and (2) the task should be open-ended enough to permit flex-
ibility and novelty in children’s responses.
The story-telling and the collage-making tasks for the pres-
ent study were developed by Amabile (1983), and have been

122

37-2-03 pages.p65 122 3/20/03, 8:35 AM

 Journal of Creative Behavior

validated in more than 30 independent studies by Amabile and

her colleagues (1983, 1996), and by others (Baer, 1991, 1993).
High inter-rater reliability and long-term validity have been
reported for these tasks (Amabile, 1983, 1996; Bear, 1994c).
For the story-telling task, each individual child was shown
the picture book, A Boy, a Dog, a Frog, and a Friend (Mayer,
1971). After looking through it to become familiar with the story,
the child was asked to tell a story in his/her own words by say-
ing one thing about each page while looking at the book’s
pictures. For the collage-making task, each child was given
identical sets of materials to work with: a 14" x 22" piece of
white tag board, a bottle of glue, and a set of over one hundred
precut pieces of construction paper in several different sizes,
shapes, and colors. Each child was asked to make an interest-
ing, silly design using all the materials provided.
The math word-problem task was adapted from Baer’s
(1991) study, where high inter-rater reliability information was
reported (Baer, 1991). For the math word-problem task, chil-
dren were asked to tell an interesting and original math word
problem. Children were not asked to solve the problem they
created, but instructed to make sure all needed information
was included so that the problem could be solved by someone
else. Children were instructed that they could use paper and
pencil to create their math problems.
All performance-based assessments were judged for their
creativity by three expert judges in each domain selected for
this study. Based on Amabile’s (1996) guidelines on consen-
sual techniques for creativity assessment, judges were selected
based on their experience and expertise in each domain. For
each domain, teachers who had specialized credentials and at
least five years of experience in teaching children’s creative
writing, art, or math were selected as judges for the study. To
avoid any biases (e.g., settings, times of the days, moods, etc.)
in scoring, all judges gathered at the same place and at the
same time for four hours and had no knowledge of the identity
of the author of each product. Each judge made his or her
assessment independently in a separate room, and rated
children’s products on a 1.0 (low) to 5.0 (high) numerical scale
based on his or her own definition of creativity. As recom-
mended by Amabile (1983, 1996), judges were not trained to
agree with one another, nor given specific criteria for creativ-
ity. However, judges were instructed to rate children’s products
relative to other children’s products of the task examined, rather
than rating them against some absolute standards. Amabile

123

37-2-03 pages.p65 123 3/20/03, 8:35 AM

 Domain-specificity of creativity in young children

suggested this is important because, the levels of creativity

produced by the ordinary children would be low in compari-
son with the greatest works produced in that domain. Once
the judgments in all performance-based assessments were
completed, the judges’ ratings on each assessment were ana-
lyzed for inter-judge reliability. Children’s scores on each assess-
ment were calculated by averaging the three judges’ ratings.
Judges were paid for their work in this study.
Procedures All the measures, except one performance-based assess-
ment, were administered individually and untimed. A small,
quiet, and simply decorated room in each participating school
was used to carry out each individually administered assess-
ment. The examiner made an effort to establish rapport with
each child in the assessment setting before the assessment
was undertaken, and all the measures were administered in a
game-like atmosphere. In most cases, all measures were ad-
ministered during one session in about 30 minutes to 1 hour
time period. All measures were administered in the same se-
quence with the Wallach-Kogan Creativity Test given first,
followed by the Real-world Divergent Thinking Test, story-
telling, and math word problem. If the child displayed some
fatigue or disinterest during the testing, testing was stopped
and administered at another time. One performance-based
assessment, collage-making, was administered in a small
group setting at a time convenient to classroom teachers. All
assessments were coded by number, not by child’s name or
child’s ID, to avoid any biases in scoring.

RESULTS Divergent Thinking Skills. The reliability of each divergent

thinking test was evaluated with inter-item correlations
(Cronbach’s alpha). Alpha coefficients were calculated for each
subtest score of the Wallach-Kogan Creativity Test and the
Real-World Divergent Thinking Test, respectively. The alpha
coefficients for the Wallach-Kogan Creativity Test and the
Real-World Divergent Thinking Test were quite high and all
adequate. Table 1 presents the alpha coefficients with the
means, standard deviations, and range of subtest scores for
the two divergent thinking tests used in the study.
Pearson correlation was used to examine the within and
between relationships of the two divergent thinking tests. Table
2 presents these results. The Wallach-Kogan Creativity Test
subtest scores, problem-solving fluency (WKF), problem-
solving originality (WKO), and problem-finding fluency
(WKPF), were strongly and significantly related to each other.

124

37-2-03 pages.p65 124 3/20/03, 8:35 AM

 Journal of Creative Behavior

TABLE 1. Mean, Standard Deviation, Range, and Alpha Coefficient for

Subtest Scores of the Wallach-Kogan Creativity Test and the
Real-World Divergent Thinking Test.
Measures M SD Range Alpha
Wallach-Kogan
Problem solving tasks
Fluency 36.60 20.20 9 – 101 .93
Originality 5.62 6.33 0 – 34 .84
Problem finding tasks 23.33 11.87 8 – 69 .77
Fluency 23.33 11.87 8 – 69 .77
Real-World
Problem solving tasks
Fluency 9.20 6.82 1 – 43 .94
Originality 1.01 2.48 0 – 13 .90
Problem finding tasks
Fluency 11.25 9.42 0 – 60 .89

Similarly, the Real-World Divergent Thinking Test subtest

scores were also significantly related to each other, except the
relationship between the problem-solving originality score
(RWO) and problem-finding fluency score (RWPF) (r = .206).
Furthermore, all subtest scores of the Wallach-Kogan Creativ-
ity Test were significantly related to the scores of the Real-
World Divergent Thinking Test.

TABLE 2. Pearson Correlations between the Wallach-Kogan Creativity

Test and Real-World Divergent Thinking Test (N = 109).
WKF WKO WKPF RWF RWO RWPF
WKF 1.00 .746* .810* .634* .449* .568*
WKO 1.00 .595* .510* .524* .331*
WKPF 1.00 .558* .365* .601*
RWF 1.00 .640* .369*
RWO 1.00 .206
RWPF 1.00
Note. WKF = WK problem-solving fluency; WKO = WK prob-
lem-solving originality; WKPF = WK problem-finding fluency;
RWF = RW problem-solving fluency; RWO = RW problem-
solving originality; RWPF = RW problem-finding fluency.
*p < 0.01

125

37-2-03 pages.p65 125 3/20/03, 8:35 AM

 Domain-specificity of creativity in young children

Creative Performances. Likewise, inter-judge reliabilities

were calculated for the story-telling, collage-making, and math
word-problem creating tasks using the Cronbach’s coefficient
alpha. As shown in Table 3, inter-judge agreement among the
three judges in each domain was fairly high in this study. In-
suring high inter-judge reliability is the most important crite-
rion of a consensual assessment procedure, since inter-judge
reliability in a study like this is equivalent to construct validity
(Amabile, 1996; Hennessey & Amabile, 1988).

TABLE 3. Mean Ratings, Standard Deviations, and Alpha Coefficient for

the Three Performance-based Assessments (N = 109).
Tasks Mean* Standard Deviation Alpha
Story-telling 2.37 1.12 .88
Collage-making 2.91 1.21 .94
Math word-problem 2.26 1.25 .92
*rating: 1 (not creative) to 5 (highly creative).

Although it was not the purpose of the study to identify “cre-

ative” children in particular domains, 40 children (37%) among
the 109 were considered “creative” (‘creative group’) in one or
more domains as judged by the ratings received for their prod-
ucts in story-telling, collage-making or the math word prob-
lem. To be inclusive (rather than exclusive), children with ratings
of 4 (creative) or 5 (highly creative) from the three judges
(mean 4.0 or greater) in any domain were regarded as ‘cre-
ative’ in the study. Fourteen (13%) children were considered
creative in story-telling, 21 (19%) in collage-making, and 16
(15%) in the math word problem. Among the 40 creative chil-
dren, seven (6%) demonstrated creativity in two domains and
only two (2%) showed creative performance in all three do-
mains. Sixteen creative children were boys and 24 were girls.
The mean age for the creative children was 7 years and 9 month
with a standardized deviation of 5 months. Sixty-nine (63%)
children in this study did not demonstrate creative perfor-
mances in any of the three domains assessed (‘non-creative
group’). Figure 1 displays the number of children who demon-
strated creative performance in one, two, or three domains.
Is Creative Pearson correlations were conducted to explore the relation-
Performance in One ships among the three performance-based assessments of
Domain Predictive
of Performance in story-telling, collage-making, and the math word-problem.
Other Domains? Among the three performance-based assessments, the only

126

37-2-03 pages.p65 126 3/20/03, 8:35 AM

 Journal of Creative Behavior

FIGURE 1. Distribution of creative children in three domains.

Creative in story –
7

3 2
2

14 2 10
Creative in Creative in

significant relationship was found between the story-telling and

the math word problem (r = .283, p = .004). However, story-
telling was not significantly related to collage-making, nor was
the math word-problem related to collage-making. Table 4
shows the correlations among the three performance-based
assessments for the 109 children.

TABLE 4. Pearson Correlations between Story-telling, Collage-making,

and the Math Word-problem (N = 109).
Story- Collage- Math
Tasks telling making problem
Story-telling 1.00 .072 .283*
Collage-making 1.00 .195
Math word-problem 1.00
*p < 0 .01

Although Table 4 displays the correlations between the three

performance-based assessments in the 109 children, it does
not reflect whether similar relationships existed for subgroups
of children who demonstrated creativity in one or more

127

37-2-03 pages.p65 127 3/20/03, 8:35 AM

 Domain-specificity of creativity in young children

domains (creative group) and children who did not show cre-
ativity in any of the three domains in this study (non-creative
group). In particular, it would be interesting to see if the overall
relationships between story-telling and math held for children
who were judged as creative.
Pearson correlations were calculated for the three perfor-
mance-based assessments in the two groups of children (cre-
ative and non-creative groups) (see Table 5). For the creative
group, no task showed a significant relationship with other
tasks. Each task showed non-significant and even negative
relationships with other tasks in different domains, indicating
creative performance in one domain was quite independent of
the creative performances in other domains. The results for
the non-creative group were rather similar to the ones for the
total sample; there was only a significant relationship between
story-telling and the math problem and there were weak, non-
significant relationships between story-telling and collage-mak-
ing and between collage-making and the math word problem.
Overall, the results indicate that creative performances in
different domains both in the creative- and non-creative groups,
as well as in the overall sample, were relatively independent of
each other. The creative group of children demonstrated stron-
ger independence between the tasks in the different domains
than the non-creative group.

TABLE 5. Pearson Correlations between the Story-telling, Collage-

making, and the Math Problem for Creative (N = 40) and Non-
Creative (N = 69) Groups of Children.

Story- Collage- Math

Tasks telling making problem
creative group
Story-telling –.203 –.022
Collage-making .045 –.148
Math problem .273* .157
non-creative group
*p < 0.05

Do General Creative Several statistical analyses were conducted to examine the

Thinking Skills
Predict Creative
relationship between the two divergent thinking tests used in
Performances? this study and the three performance-based assessments. First,
Pearson correlation analyses were conducted to examine the

128

37-2-03 pages.p65 128 3/20/03, 8:35 AM

 Journal of Creative Behavior

relationships between the Wallach-Kogan Creativity Test

(WK), Real-World Divergent Thinking Test (RW), and the
three performance-based assessments. Table 6 presents the
detailed results.
Two of the six subtest scores of the two divergent thinking
tests were significantly related to the story-telling task (p < .01).
The highest correlation existed for the Wallach-Kogan prob-
lem-solving originality score (WKO) (r = .365). No significant
relationship was found between any of the subtest scores of
the two divergent thinking tests and the math word-problem
task or collage-making tasks.

TABLE 6. Pearson Correlations for the Wallach-Kogan Creativity Test,

Real World Divergent Thinking Test, and Three Performance-
based Assessments (N = 109).

Measures Story-telling Collage-making Math problem

WKF .306* .069 .120
WKO .365* .150 .192
WKPF .231 .073 .135
RWF .243 –.011 .026
RWO .247 .038 .019
RWPF .157 .197 .116
Note. WKF = WK problem-solving fluency; WKO = WK prob-
lem-solving originality; WKPF = WK problem-finding fluency;
RWF = RW problem-solving fluency; RWO = RW problem-
solving originality; RWPF = RW problem-finding fluency
*p < 0.01

Multiple regression analyses were also conducted to exam-

ine how much of the variances for the three performance-based
assessments were explained by the six subtest scores of the
Wallach-Kogan and the Real-World Divergent Thinking Tests.
Examination of the regression analysis reveals that the six
subtest scores of the two divergent thinking tests did not ex-
plain significant nor substantial proportions of the variances
(at p < .01 level) in any of the three performance-based assess-
ments. The results indicated, however, that the six subtest
scores of the two divergent thinking tests accounted for a mar-
ginally significant 14% (p = .02) of the variance in story-telling

129

37-2-03 pages.p65 129 3/20/03, 8:35 AM

 Domain-specificity of creativity in young children

task. The six subtest scores accounted for only 8% of the vari-
ance in the collage and 5.7% in the math word-problem tasks.
The results were similar when multiple regression analyses
were separately conducted for each divergent thinking test
(Han, 2000).
Although the multiple regression analysis provides informa-
tion about the extent to which the Wallach-Kogan Creativity
Test and the Real-World Divergent Thinking Test explain and
predict each dependent variable (performance-based assess-
ments) separately, the procedure ignores the very essence and
richness of a multi-faceted phenomenon between the multiple
dependent and independent variables. Canonical correlations,
therefore, were used to demonstrate an interrelationship be-
tween these two sets of multiple variables. Canonical analysis
was conducted to explain the extent to which one set of crite-
rion variables (story-telling, collage-making, and math-word
problems) were predicted or explained by another set of pre-
dictor variables (the six subtest scores of the two divergent
thinking tests). Results indicated that the predictor variate was
not significantly correlated with the criterion variate [Rc = .407,
χ2 (100) = 24.23, p = .148]. Any combination of the six subtests
did not explain or predict significantly any combination of the
three performance-based assessments. The divergent think-
ing measures and the creative performances in three domains
were independent of one another. Since no significant correla-
tion was found between the two sets of variables, no further
analysis was made.

STUDY 2 In study 1, the results of the study have been described in

the form of group and quantitative data. However, such an
approach has sacrificed some individual qualitative data. Study
2 describes the results and profiles of three individual highly
creative children (names changed to protect their identity),
hoping to provide enriching and qualitative specification to the
quantitative results presented in study 1. Study 2 is built on
several assumptions: (1) it might not be possible to generalize
about creativity because each creative child is too unique to
be generalized; (2) quantitative approach to creativity may
prevents or prohibit serious examination of the individual; there-
fore, (3) creativity in young children may be better understood
via careful observation of individual creative young child in a
particular domain. Overall, this study makes an effort to take
a grasp of how each individual young child shows his/her cre-
ativity in a particular domain.

130

37-2-03 pages.p65 130 3/20/03, 8:35 AM

 Journal of Creative Behavior

Mindi: The most Mindi was eight years and one month old when she was
creative artist
tested. Her parents are from Taiwan, though Mindi was born in
America. Mindi had the highest mean rating (5.00) on the col-
lage-making task. All three judges selected Mindi’s collage as
the most creative among those of 109 children. Mindi’s col-
lage is presented in Figure 2. While other children used only
about half of the provided papers, she used nearly all of them.
Rather than using the papers as provided, she made an effort
to change the shapes by ripping, bending, folding, or overlap-
ping the papers. The dominant creative characteristic of Mindi’s
collage was elaboration. According to the three art judges,
Mindi’s collage revealed a very exceptional talent in art. Sub-
sequent interaction with Mindi’s teachers revealed that Mindi
has won first prize for her drawings at least three times in
national and state art contests.

FIGURE 2. Mindi’s collage.

In contrast to her superb artistic ability, however, Mindi

showed an average performance in the story-telling task.
Mindi’s story about the book, A boy, a dog, a frog, and a
friend (Mayer, 1971) was short and did not include any inter-
esting conversation, names of characters, or difficult vocabu-
laries. The language judges indicated Mindi’s story was only a
description of pictures rather than a story. The mean rating for
Mindi’s story-telling from the three judges was 1.33, and it was
about one standard deviation below the group mean of 2.37.
Mindi’s story-telling is reproduced in Figure 3.

131

37-2-03 pages.p65 131 3/20/03, 8:35 AM

 Domain-specificity of creativity in young children

FIGURE 3. Mindi’s story on the book, A boy, a dog, a frog, and a friend
(Mayer, 1971).

The little boy goes out fishing. And then he thinks he

caught something very, very big. And this thing pulled
him inside the water. And it was a turtle. And the boy
came running after it. Then the boy’s dog was mad at it,
but that made the turtle mad at it too. So it bit the dog’s
paw. Then the boy tries to pull the dog so the turtle will let
go, but it didn’t so he just carried the turtle too. Then on
the way back the turtle dropped. Then the boy was kind
of happy. He was starting to go home. Pack up. Then the
turtle bites the dog’s tail again. Then he pulls him into
the water. Then the boy wants to go swim. The dog al-
ready swimmed out. So he dressed up again. But then
the turtle kind of plays dead. Then the boy kind of feels
sorry for it. So he fishes him back to land. They kind of
carry him to a place to bury him. But the turtle was alive.
And XXX the dog. And the boy was happy it was alive.
And then they went home.

Mindi’s performance on the math word-problem was not

rated exceptionally high either. The mean rating from the three
judges was 2.33. The rating was similar to the group mean of
2.26. Mindi’s math word-problem follows:
There is a boy. He’s hungry. And there are twenty
apples on the tree. He picks six off, then he eats four of
them. How many are on the tree and the ones that he
didn’t eat?

Mindi’s scores on the two divergent thinking tests were simi-

lar to group means for both tests, although her problem-find-
ing score on the Real-World Divergent Thinking Test (RWPF)
was higher than the group mean. Table 7 shows Mindi’s per-
formances on all subtests on the Wallach-Kogan Creativity
Test and the Real-World Divergent Thinking Test.
In sum, Mindi was the most creative artist in the present
study. However, Mindi did not demonstrate any creative perfor-
mances in other tasks in different domains. Mindi’s divergent
thinking skills did not explain her exceptional artistic creative
performance in this study.

132

37-2-03 pages.p65 132 3/20/03, 8:35 AM

 Journal of Creative Behavior

TABLE 7. Mindi’s, Albert’s, and Hanna’s performances on the Wallach-

Kogan Creativity Test and the Real-World Divergent Think-
ing Test with the Group Means.
Wallach & Kogan Real-world DT
WKF WKO WKPF RWF RWO RWPF
Mindi 34 6 25 5 0 19
Albert 29 8 22 9 1 13
Hanna 101 18 53 21 2 46
Group
mean 36.60 5.62 23.33 9.20 1.01 11.25

Albert: A creative Albert was eight years and three months old when he was
story-teller.
tested. Albert is one of the five children who received a mean
rating of 5.00 from the three judges for the story-telling task.
Albert’s story demonstrated novelty of word choice, richness
of imagery, sophisticated expression, rhythm, humor, and a
well organized and planned plot throughout the story. Albert’s
story is reproduced in Figure 4.

FIGURE 4. Albert’s story on the book, A boy, a dog, a frog, and a friend
(Mayer, 1971).
Once upon a time a boy went to a lake to find a XXX*
fish. His friends, a dog and a frog sat near him to watch
him. When the fishing rod tugged, the boy knew that he
had a bite. He began to pull this up. The dog and the frog
began to watch him. He tugged so hard and so was the
boy. Everything. He pulled and pulled but the fish wouldn’t
come out. The dog began to growl. And the frog looks
like he is about to leave. Then the boy splashed in, still
unlet go fishing rod. The dog and the frog jumped after
him. Turned out it was only a turtle. The turtle walked
away with the fishing rod hook. The frog and the dog
began to swim after him, followed by the boy. The dog
wanted to fight. He began to growl and showed his mighty
teeth. But the turtle turned out to be a snapping turtle. It
just bit the dog. And he moaned. But now the boy was at
shore. The frog leaped away. The boy pulled and pulled.
And the turtle pulled and pulled. And so boy began to
carry the dog. And the turtle snapped and bounced it.
Then the turtle let go and fell off. The dog was happier.
So did the boy. They are pulled themselves to shore. Boy

133

37-2-03 pages.p65 133 3/20/03, 8:35 AM

 Domain-specificity of creativity in young children

sat down his things and he began to count his fish. But
then along came the turtle again. He bit the dog’s tail
and the boy dropped his things. The turtle began to carry
the dog away. “Oh, No!” shouted the boy. “Oh, No!”
hopped the frog. The boy undressed himself and then he
went into to get the dog. And the frog got upon a lily pad
waiting for the boy. When the boy had got the dog, he
dressed himself again. Along came the turtle. Only the
turtle looked happier. He did not snap. But the boy had
to look at it very carefully. He was dead! The frog was sad
and the dog was sad. But the boy was mad at the dog.
The boy took the turtle to a grassy part near the lake.
The dog felt small because he felt guilty while turtle lay
down. But then the turtle opened his eyes. And the boy
was picking the flower from near the XXX* river. But the
turtle was about to hit the boy’s fishing rod. The dog and
frog had seen it, and then finally the boy. The boy was
very happy and so was the turtle. He loved the fishing
rod. The boy, the turtle, the frog and the dog went home
happy, especially the boy. He found a new person to carry
his fishing rod.
Note. * Albert made some unique names for the fish
and the river. But they were unrecognizable during the
transcription.

In contrast to his exceptional performance in story-telling,

Albert’s performances in collage-making and the math word-
problem were not very impressive. Although he showed a simi-
lar level of task-commitment throughout all tasks, his collage
and math problem were rated relatively low. The mean rating
for his collage from the three judges was 1.67. His collage did
not display any novel ideas in using and organizing the materi-
als, variation in shapes, complexity or silliness.
Albert’s math word-problem was rated only 1.33 by the three
judges. His problem was very short and simple. Albert did not
include any complexity, difficult calculations, high numbers,
or a high level of reasoning in the problem. In fact, Albert’s
problem could not be solved because it had some missing
numbers that would be necessary to solve the problem. Albert’s
math problem was as follows:
There were thirty-four frogs and cats. Ten of the cats
ate frog. How many frogs are left?

134

37-2-03 pages.p65 134 3/20/03, 8:35 AM

 Journal of Creative Behavior

Albert was not a highly divergent thinker either. Albert’s

performances both on the Wallach-Kogan Creativity Test and
the Real-World Divergent Thinking Test were close to the
group means. Albert’s performances on the two divergent think-
ing tests are reported in Table 7.
Hanna: The most Hanna was seven years and nine months old when she was
highly divergent tested. The most dominant creative characteristics for Hanna
thinker.
were her fluency and flexibility. She received the highest score
on the Wallach-Kogan Creativity Test and the second high-
est score on the Real-World Divergent Thinking Test among
the 109 children in the present study. For example, she gave
101 different responses on the Wallach-Kogan problem-solv-
ing fluency (WKF) subtest; the mean score for the sample in
the study was 36.60. Hanna’s score on the WKF was more
than three standard deviations above the mean (see Table 7).
Hanna provided very many ideas for both the problem-solv-
ing and the problem-finding subtests, and on both the verbal
and the nonverbal tests of the Wallach-Kogan Creativity Test
and the Real-World Divergent Thinking Test. However, the
most interesting responses from Hanna were the ones on the
problem-finding task on the nonverbal subtest (pattern mean-
ing) of the Wallach-Kogan Creativity Test. When she was pro-
vided a blank piece of paper and asked to make a pattern (or
line) of her own and tell all the different things it could be, she
drew an interesting pattern and generated lots of interesting
explanations for it. The pattern made by Hanna and her re-
sponses are reproduced in Figure 5.

FIGURE 5. Hanna’s problem-finding responses (Pattern Meaning) on the

Wallach-Kogan Creativity Test.

Responses: 1) path 2) street 3) maze 4) side walk in a small

village 5) two stepping rocks in a stream 6) two papers held
by a person 7) two boxes that I am trying to fix it back 8) one
leg and another leg is outside 9) bridge that connect two houses

135

37-2-03 pages.p65 135 3/20/03, 8:35 AM

 Domain-specificity of creativity in young children

10) two hands pulling somebody’s hair 11) diving board 12)
shades 13) two magnifying glasses 14) piece of paper and
someone is going to write on it 15) measuring machine 16)
telescope 17) two faces looking each other 18) a part of a build-
ing 19) two hands trying to open something 20) two walls to
be squashed 21) a square inside (there are three squares all)
22) something that I am trying to tear out bad 23) intersection

Hanna’s performances on both divergent thinking tests were

quite exceptional for a seven-year-old girl. However, her perfor-
mances on the story-telling, math word-problem, and collage-
making tasks were not very impressive. The judges’mean
ratings for Hanna’s story-telling, math word-problem, and col-
lage-making tasks were 2.00, 1.67, and 3.33 respectively. The
performances in language and math were below the mean,
and her performance in art was a little higher than the group
mean. It is possible that Hanna may demonstrate creative abil-
ity in other domains (i.e., music, science, inter-personal) or on
different tasks in the three domains of language, art, and math.
However, it was interesting to see that the most divergent thinker
in the study did not demonstrate any creative performance in
any of the areas tested.
In summary, these three children demonstrated creative
performance in a particular area. These children were very

FIGURE 6. Individual differences addressed in different creative perfor-

mances.

136

37-2-03 pages.p65 136 3/20/03, 8:35 AM

 Journal of Creative Behavior

similar in that they were independent, self-directed, and persis-

tent on the various tasks used in the study. However, they were
also very different in the kinds and levels of creative perfor-
mances they revealed. Each of them had a dominant, strong,
or special area in which he or she displayed his/her creativity.
Creativity profiles for Mindi, Albert, and Hanna were expressed
in very divergent ways, and would be best explained as being
domain-specific. Figure 6 shows that there is considerable
intra-individual variation in creative ability by domain in Mindi,
Albert, and Hanna.

DISCUSSION The results of the study suggest that there is a considerable

Creativity as intra-individual variation in creative ability by domain in the
Domain-Specific
109 second grade children who participated in the study. As
exemplified in the case studies of Mindi, Albert, and Hanna,
many children in the study exhibited a range of creative abili-
ties across different domains rather than a uniform creative
ability in diverse domains. It is implied from the study that it is
hard to reliably predict a child’s creative ability in one domain
based on his/her creative ability in other domains, thereby pro-
viding some empirical support for the domain-specific theory
of creativity.
Specifically, the negative and non-substantial correlations
found among the performance-based assessments in three
different domains for the creative group of children in the
study might serve as a major source of evidence for creativity
as domain-specific. The fact that only 2 out of 109 children
demonstrated creative performances in all three domains
also makes a strong and reasonable case for the position of
domain-specificity of creativity.
Although a significant relationship existed between story-
telling and math word problem tasks in the whole sample of
109 children, the correlation between the tasks was not very
high (r = .283). This might be explained in the fact that both
tasks utilized a similar task format. In the story-telling task,
children were asked to generate a story based on a word-less
picture book, and in the math word problem, children were
asked to make up a math-related story problem. The fact that
both tasks required verbal explanation might be responsible
for the result. The result also might be attributed to individual
“style” that is defined as a manner of approaching and accom-
plishing tasks (Miller, 1991). As Adams (1993) suggested, the
weak but significant correlation between such tasks may re-
flect the systematic influence of style rather than links between

137

37-2-03 pages.p65 137 3/20/03, 8:35 AM

 Domain-specificity of creativity in young children

the content areas themselves. Adams indicated that if individu-

als adopt the same style in two different tasks, the positive and
significant correlation between the tasks is the function of style,
not a true reflection of association between different knowl-
edge domains. However, whether the relationship between t
he story-telling and math word problem tasks is attributed to
domain-general creative ability, similar task format, individual’s
style toward a task, or other affective factors that may cut across
domains — such as creative self-efficacy or motivation — can
not be explained empirically in this study. Replication would
be useful in examining whether similar patterns are found in
further studies and in exploring possible causes for associa-
tion between the tasks.
Taken together, these findings are in contrast to the previ-
ous research (e.g., Hocevar, 1980; Hong, Milgram & Gorsky,
1995) that suggested that creative performances in different
domains for young children were significantly related to each
other and rather domain-general. The results also stand out
against the commonly held claim that children are creative in
many different domains. These findings are, then again, con-
sistent with some other earlier research (e.g., Baer, 1991) that
suggested creative performances are domain-specific. Never-
theless, since to-date no study (including this one) has found
absolute independence or dependence between tasks in dif-
ferent domains, interpretation and conclusion regarding
creativity’s domain-generality and -specificity should be made
very carefully.
Divergent Thinking One of the most striking findings of the study was that
as a General
Creative Thinking
divergent thinking measures did not have great power in pre-
Ability dicting creative performances in at least two of three, if not all,
domains assessed in the study. As addressed in the case of
Hanna, even the most divergent child in the study did not nec-
essarily show creative performances in particular areas. Nei-
ther of the divergent thinking tests administered in the study,
in separate or in a battery, predicted real creative behaviors of
109 young children. The amount of variances explained by the
two divergent thinking tests in the collage and math tasks were
relatively small. Although two subtests of the Wallach-Kogan
Creativity Tests were significantly related to the story-telling
performance, the amount of variance explained by the two
divergent measures, in separate or in a battery, was not sub-
stantial. In creativity research, ‘good’ test correlations with
outside validating criteria run between .40 and .65 because of
the host of factors affecting test and criterion data (Bartlett &

138

37-2-03 pages.p65 138 3/20/03, 8:35 AM

 Journal of Creative Behavior

Davis, 1974). None of the correlations between the two diver-

gent thinking tests and the story-telling task reached this range
of “good”. More importantly, the canonical correlation indicated
the two divergent thinking tests and children’s creative perfor-
mances in the three domains were independent of one another.
The findings of the study both complement and contrast
with earlier research. These findings are consistent with some
of previous research in that divergent thinking measures were
often weakly but significantly related to creative activities in
the language domain (i.e., story-telling, writing) but not to cre-
ative behaviors in other domains (Baer, 1991; Runco, 1986),
implying a possible verbal bias in divergent thinking tests.
The results also contrast with earlier research efforts (i.e.,
Bartlett & Davis, 1974; Hocevar, 1980; Hong, Milgram, &
Gorsky, 1995; Torrance, 1972; Wallach & Wing, 1969; Wallach
& Kogan, 1965) that have found significant relationships be-
tween divergent thinking measures and diverse creative per-
formances (i.e., leadership, art, social service, literature, writing,
math, science, crafts). As indicated, the contrast might be due
to the fact that the earlier research has depended on self-
report scales to measure children’s creative performances in
different domains. As indicated, self-report scales, however,
have been criticized for their questionable validity and the
response-set bias that may lead individuals to systematically
underestimate or over-estimate their creative behaviors across
all domains (Baer, 1998, 1999). It was also interesting the find
out that the Real-World Divergent Thinking Test was not more
predictive of creative performances than a conventional diver-
gent thinking test like the Wallach-Kogan. The results do not
support the findings of previous research (Runco & Okuda,
1988) and have raised a question about the usefulness of the
real-world divergent thinking test in predicting creative perfor-
mances in young children.
Overall, these findings of the present study suggest the lack,
if not absence, of general creative thinking skills in explaining
children’s creative performances. On the other hand, the
results might be interpreted to say that divergent thinking mea-
sures may not represent a general creative thinking ability
appropriately. These findings have important practical impli-
cations, since divergent thinking tests are widely used to iden-
tify creative children in a wide variety of domains.
In conclusion, the findings of this research support the
position that creative ability in young children is rather (but
not absolutely) domain-specific. Children’s creative behaviors

139

37-2-03 pages.p65 139 3/20/03, 8:35 AM

 Domain-specificity of creativity in young children

were expressed in widely divergent ways, implying that “diverg-

ing” from divergent thinking tests is called for. A domain-
specific view of creativity offers a more useful and construc-
tive knowledge about a child’s strength than the domain-
general view which mostly utilizes divergent thinking measures
to identify a general and domain-transcending capacity of cre-
ativity. This study alludes to the fact that understanding and
identifying creative abilities in our young children is far more
complicated and complex than we have believed.

REFERENCES ADAMS, M. (1993). An empirical investigation of domain-specific theories

of preschool children’s cognitive abilities. Unpublished doctoral
dissertation. Tufts University.
AMABILE, T. M. (1983). The social psychology of creativity: A componential
conceptualization. Journal of Personality and Social Psychology, 45
(2), 357-376.
AMABILE, T. M. (1996). Creativity in context. Boulder, CO: Westview Press.
ANASTASI, A. (1982). Psychological testing. New York: Macmillan.
BAER, J. (1991). Generality of creativity across performance domains.
Creativity Research Journal, 4, 23-39.
BAER, J. (1993). Creativity and divergent thinking: A task-specific
approach. Hillsdale, NJ: Lawrence Erlbaum Associates.
BAER, J. (1994a). Divergent thinking is not a general trait: A multi-domain
training experiment. Creativity Research Journal, 7, 35-46.
BAER, J. (1994b). Why you shouldn’t trust creativity tests. Educational
Leadership, 51 (4), 80-83.
BAER, J. (1994c). Performance assessment of creativity: Do they have long
term stability? Roeper Review, 17 (1), 7-11.
BAER, J. (1998). The case for domain specificity of creativity. Creativity
Research Journal, 11 (2), 173-177.
BAER, J. (1999). Domains of creativity. In M. A. Runco & S. R. Pritzer (Eds.),
Encyclopedia of creativity Vol. 1 (pp. 591-596). San Diego, CA:
Academic Press.
BARRON, F. (1988). Putting creativity to work. In R. J. Sternberg (Ed.), The
nature of creativity (pp. 76-98) New York: Cambridge University Press.
BARTLETT, M. M., & DAVIS, G. (1974). Do the Wallach-Kogan Creativity
Tests predict real creative behavior? Perceptual and Motor Skills, 39,
730.
BROWN, R. T. (1989). Creativity: What are we to measure? In J. A. Glover, R.
R. Ronning, & C. R. Reynolds (Eds.), Handbook of creativity. (pp. 3-
32). New York: Plenum Press. Bobbs-Merrill.
CRAMOND, B. (1994). We can trust creativity tests. Educational
Leadership, 52 (2), 70-71.
CSIKSZENTMIHALYI, M. (1988). Society, culture, and person: A system view
of creativity. In R. J. Sternberg (Ed.), The nature of creativity (pp. 325-
339). New York: Cambridge University Press.
CSIKSZENTMIHALYI, M. (1990). The domains of creativity. In M. A. Runco
& R. S. Albert (Eds.), Theories of creativity (pp. 190-214). London: Sage.

140

37-2-03 pages.p65 140 3/20/03, 8:35 AM

 Journal of Creative Behavior

FELDHUSEN, J. F., & GOH, B. E. (1995). Assessing and accessing creativity:

An integrated review of theory, research, and development. Creativity
Research Journal, 3 (3), 231-247.
FELDMAN, D. H. (1980). Beyond universals in cognitive development.
Norwood, NJ: Ablex.
FELDMAN, D. H. (Ed.). (1982). New directions for child development:
No. 17. Developmental approaches to giftedness and creativity.
Sanfransisco: Jossey-Bass.
FELDMAN, D. H. (1994). Beyond universals in cognitive development.
(2nd Ed.). Norwood, NJ: Ablex.
GARDNER, H. (1983). Frames of minds. New York: Basic Books.
GARDNER, H. (1993a). Multiple intelligences. New York: Basic Books.
GARDNER, H. (1993b). Creating minds. New York: Basic Books.
GUILFORD, J. P. (1967). The nature of human intelligence. New York:
McGraw-Hill.
HAN, K. S. (2000). Varieties of creativity: Investigating the domain-
specificity of creativity in young children. Unpublished doctoral
dissertation. University of Nebraska.
HENNESSEY, B. A., & AMABILE, T. M. (1988). Story-telling: A method for
assessing children’s creativity. Journal of Creative Behavior, 22 (4),
235-246.
HOCEVAR, D. (1980). Intelligence, Divergent thinking and creativity.
Intelligence, 4, 25-40.
HONG, E., & MILGRAM, R. M. (1991). Original thinking in preschool children:
A validation of ideational fluency measures. Creativity Research
Journal, 4 (3), 253-260.
HONG, E., MILGRAM, R. M., & GORSKY, H. (1995). Original thinking as a
predictor of creative performance in young children. Roeper Review,
18 (2), 147-149.
KOGAN, N. (1994). Diverging from divergent thinking. Contemporary
Psychology, 39 (3), 291-292.
MAYER, M. (1971). A boy, a dog, a frog, and a friend. New York: Dial Books
for Young Readers.
MILGRAM, R. M. (1990). Creativity: An idea whose time has come and gone?
In M. A. Runco., & R. S. Albert. (Eds.), Theories of creativity. London:
Sage.
MILGRAM, R. M., & MILGRAM, N. A. (1976). Creative thinking and creative
performance in Israeli children. Journal of Educational Psychology,
68 (3), 255-259.
MILLER, A. (1991). Personality types: A modern synthesis. Calgary Alberta,
Canada: University of Calgary Press.
OKUDA, S. M., RUNCO, M. A., & BERGER, D. E. (1991). Creativity and the
finding and solving of real-world problems. Journal of
Psychoeducational Assessment, 9, 45-53.
PLUCKER, J. (1998). Beware of simple conclusions: The case for content
generality of creativity. Creativity Research Journal, 11 (2), 179-182.
RUNCO, M. A. (1986). Divergent thinking and creative performance in gifted
and nongifted children. Educational and Psychological Measurement,
46, 375-384.

141

37-2-03 pages.p65 141 3/20/03, 8:35 AM

 Domain-specificity of creativity in young children

RUNCO, M. A. (1987). The generality of creative performance in gifted and

nongifted children. Gifted Child Quarterly, 31 (3), 121-125.
RUNCO, M. A. (1989). The creativity of children’s art. Child Study Journal,
19, 177-190.
RUNCO, M. A. (1993). Divergent thinking, creativity, and giftedness. Gifted
Child Quarterly, 37 (1), 16-22.
RUNCO, M. A. & NEMIRO, J. (1994). Problem finding, creativity, and
giftedness. Roeper Review, 16 (4), 235-240.
RUNCO, M. A., & OKUDA, S. M. (1988). Problem discovery, divergent
thinking, and the creative process. Journal of Youth and Adolescence,
17 (3), 211-220.
TARDIF, T. Z., & STERNBERG, R. J. (1988). What do we know about
creativity? In R. J. Sternberg (Ed.), The nature of creativity (pp. 429-
440). New York: Cambridge University Press.
TORRANCE, E. P. (1966). Torrance Tests of Creative Thinking. Bensenville,
IL: Scholastic Testing Service.
TORRANCE, E. P. (1972). Predictive validity of the Torrance Tests of Creative
Thinking. The Journal of Creative Behavior, 6 (4), 236-253.
TORRANCE, E. P. (1988). Creativity as manifest in testing. In R. J. Sternberg.
(Ed.). The nature of creativity (pp. 43-75). New York: Cambridge
University Press.
WAKEFIELD, J. F. (1985). Toward creativity: Problem finding in a divergent-
thinking exercise. Child Study Journal, 15, 265-270.
WAKEFIELD, J. F. (1992). Creative thinking: Problem solving skills and
arts orientation. Norwood, NJ: Ablex.
WALLACH, M. A., & KOGAN, N. (1965). Modes of thinking in young
children. New York: Holt, Rinehart & Winston.
WALLACH, M. A., & WING, C. W. (1969). The talented students: A
validation of the creativity-intelligence distinction. New York: Holt,
Rinehart & Winston.

Ki-Soon Han, Ph.D., University of Incheon, Department of Education, 177

Dowha-Dong, Nam-Gu, Incheon, Korea 402-749, Tel: +82-32-770-8168, HP:
+82-11-9903-2306, Fax:+82-32-777-1645, Email: han@incheon.ac.kr

142

37-2-03 pages.p65 142 3/20/03, 8:35 AM