Influences of Social and Educational Environments On Creativity During Adolescence: Does SES Matter?
Influences of Social and Educational Environments On Creativity During Adolescence: Does SES Matter?
Influences of Social and Educational Environments On Creativity During Adolescence: Does SES Matter?
David Yun Dai , Xiaoyuan Tan , Deepti Marathe , Anna Valtcheva , Robert M.
Pruzek & Jiliang Shen
To cite this article: David Yun Dai , Xiaoyuan Tan , Deepti Marathe , Anna Valtcheva , Robert
M. Pruzek & Jiliang Shen (2012) Influences of Social and Educational Environments on Creativity
During Adolescence: Does SES Matter?, Creativity Research Journal, 24:2-3, 191-199, DOI:
10.1080/10400419.2012.677338
Jiliang Shen
Beijing Normal University
It is well established that there is an academic achievement gap between students from
high and low socioeconomic family backgrounds. However, how being brought up
and living in different socioeconomic backgrounds impacts adolescent development,
particularly their creative capabilities and creativity-related personality traits, is not well
understood. This study compared creative capabilities and traits of 8th grade students of
2 school districts: a suburban, upper-middle class community and an urban community
with a large proportion of families under poverty, located in a northeastern state. The
results provide compelling evidence for a creativity gap. The study also found a possible
mediating role of academic achievement and intrinsic cognitive motivation, suggesting
that the psychosocial processes and mechanisms leading to the creativity gap are trac-
table. Implications of the findings for optimal adolescent development and social inter-
ventions are discussed.
It is well established that there is an academic achieve- research focused on personality factors in relation to
ment gap between students from high and low socioeco- creativity, whereas empirical research on the SES-related
nomic family backgrounds (Byrnes, 2003; Frederickson factors on the development of creativity is sparse and
& Petrides, 2008; McLoyd, 1998; Murdock, 2000). sporadic, most conducted back in 1980s or earlier. Lim-
Multiple factors, mechanisms, and pathways through ited evidence points to a distinct advantage enjoyed by
which low SES puts students at risk for academic failure children of higher SES over children of low SES, not only
have also been researched in depth (Grissmer, William- in academic achievement but in creative thinking as well
son, Kirby, & Berends, 1998). They involve, among (Lichtenwalner & Maxwell, 1969; Milgram, 1983;
others, parental supervision, material resources, and Ogletree, 1971); this phenomenon seems prevalent across
intellectual stimulation and value. cultures (e.g., Vijayalakshmi, 1980). Most studies
The question of a possible creativity gap between high traced this difference to early child-rearing practices
and low socioeconomic status (SES) or, more broadly, of and parenting styles (Harrington, Block & Block, 1987;
social and educational underpinnings of SES underlying Lichtenwalner & Maxwell, 1969), however, findings are
the development of creative potential, has not been not conclusive (see Miller & Gerard, 1979).
brought up and well addressed. One reason for the lack In a more recent study, Niu (2007) explored the inter-
of research attention is that creativity is often conceptua- play of individual factors (intelligence, personality,
lized and researched as an individual differences variable motivation, thinking styles and knowledge) and environ-
(e.g., Prabhu, Sutton, & Sauser, 2008). A preliminary mental factors (family and school) on creativity among
search in the PsycInfo database found a bulk of empirical Chinese students. After controlling for intelligence, Niu
found that social factors alone predicted as much vari-
Correspondence should be sent to David Yun Dai, ED233 ance in creativity as all the individual factors combined.
Educational Psychology, University at Albany, SUNY, Albany, NY In an intervention study, Tsao (2008) found that guided
12222. E-mail: ddai@uamail.albany.edu
192 DAI ET AL.
play in the classroom promoted children’s creativity in but can be enhanced (or hindered) by one’s social and
writing and conversation (see also Moore & Russ, educational environments during adolescence.
2008). In a longitudinal study, Gottfried, Gottfried, Particularly relevant to this study are different home
Bathurst, and Guerin (1994) found that children’s intrin- environments and educational experiences in school as
sic motivation was related to parental encouragement for perceived by students from high and low SES back-
task endogeny. Together, this body of research suggests a grounds. This study posited that parents with higher
set of psychosocial factors potentially mediating SES and levels of education will more likely support and engage
creativity. their adolescent children in a variety of intellectual activi-
Another reason for the lack of research on SES and ties and encourage the expression of those personal char-
creativity is that standard, reliable measures of acteristics conducive to creativity. One hypothesis was
achievement are easier to obtain than those of creativity. that the ethos and resources of a high-SES neighborhood
Theoretically, creativity can be measured in multiple and school district are more likely to be supportive of
ways, with different psychological underpinnings. For explorative, intellectually stimulating activities. There-
example, creative thinking can be operationalized in fore, the study selected two school districts that have
many ways, such as thinking divergently (Torrance, contrasting SES, but otherwise are comparable. Along
1972), constructing and combining broad categories the same line of logic, the study included several psycho-
(Cropley, 2000), or formulating and working with para- social measures such as how students perceived the
doxical ideas (Rothenberg, 1979). Also open to debate social-contextual support for individuality and creative
is the issue of the extent to which creativity can be seen expressions and how intrinsically motivated they are to
as an ability dissociated from experience and knowledge. pursue cognitive tasks.
Runco, Dow, and Smith (2006) compared two types of
performance: standard divergent thinking tasks (e.g.,
‘‘List uses for a shoe’’) and more knowledge-based idea- RESEARCH QUESTIONS
tion. They found evidence for a distinct knowledge basis
for creative ideation. Mumford, Marks, Connelly, Zec- The following research questions were addressed in the
caro, and Johnson (1998) also found a domain-specific study:
correlate of divergent thinking. Evidence seems to sug-
gest an experiential and knowledge basis of creative idea- 1. Are there major differences in divergent thinking
tion. Nevertheless, when divergent thinking performance and creativity-related personality traits between
of children was used to predict adult real-life creative students of the two school districts, which feature
achievement, it demonstrates higher predictive validity contrasting socio-economic characteristics?
than a measure of intelligence (Cramond, Matthews- 2. Do parent education levels make a difference
Morgan, Bandalos, & Zuo, 2005; Plucker, 1999). despite the fact their children live in a low-SES
An interest in SES and creativity put stakes on deve- community? Do school effects compensate for
lopmental and educational underpinnings of creativity low SES family effects?
during adolescence (Dai & Shen, 2008). In keeping with 3. Do parent education, school (and neighborhood)
the research tradition, creativity in this study was type, academic achievement (grades), perceived
defined and assessed in two broad categories: divergent parent and teacher support for the expression of
thinking and personality traits closely related to creative personal characteristics conducive to creativity,
expressions. Divergent thinking is the ability to and creative inclinations predict adolescent cre-
generate many ideas that are appropriate to the task ative capabilities as measured by a set of divergent
at hand (fluency), and flexible in crossing the boundary thinking tasks?
of categories (fluency) and original in terms of rarity
(originality). Divergent thinking is enhanced not only
METHODS
through cognitive development, but also through an
education that supports an enriched, inquisitive, open
Participants
mind. Although divergent thinking is assessed using per-
formance measures, creative traits are defined as a set of A total of 234 eighth-grade students participated in the
motivational, attitudinal, and behavioral tendencies or study in the spring semester of their academic year.
proclivities conducive to creative ideation and pro- One hundred and twenty six came from an urban middle
ductivity, and typically measured by self-report instru- school and 108 from two middle schools in a suburban
ments. The underlying assumption regarding the school district, all located in a northeastern state. The
relationship between SES and creativity is that traits two school districts are comparable in school size (an
such as openness to experience and intrinsic cognitive average of 508 vs. 541 students), the number of teachers
motivation are likely nurtured early on in one’s life, (39 vs. 41), limited English proficiency (0.6% vs. 2.2%),
SOCIAL/EDUCATIONAL ENVIRONMENTS AND CREATIVITY IN ADOLESCENCE 193
and ethnic makeup (90.2% vs. 87.1% White), but different of categories the responses fall into; 14 broad categories
in the number of students eligible for free lunch (5.1% vs. were identified for the planet question, 21 for the toy dog
40.3%) and parental education levels (80% vs. 18% gradu- question, 21 for the parallel lines question, and 16 for the
ate degrees, and 4% vs. 50% not having college edu- dividing a square problem (see Appendix A for the names
cation). The records of 5 students were excluded from of categories for the former three items); and (c) orig-
data analyses due to a large number of missing values inality, based on how frequently a specific response was
for each of the cases, resulting in a final sample of 229 given by the total sample. Two points were given for a
students. response of less than 5% chance of occurrence, and 1
point for a response of 5–10% chance.
parental support for adventurousness and openness to forming the six variables because their inclusion
experience states that ‘‘My parents encourage me to try would make interpretation of the factors conceptually
out new things, even though it may lead to failure.’’ An difficult.
item tapping into teacher support for intrinsic task For the divergent thinking test scores, the within-item
motivation states that ‘‘When I am engrossed in solving intercorrelations of fluency, flexibility, and originality
and thinking about a problem, my teachers encourage and across-item correlations among four measures of
me to keep going even when it delays my other work.’’ fluency, flexibility, and originality were examined. As
A 5-point Likert scale was used, ranging from hardly ever within-item intercorrelations of the three indexes (on
(1) to almost always (5). average .72 for planet, .92 for parallel lines, .79 for square,
and .81 for toy dog) were higher than across-item inter-
correlations of fluency, flexibility, and originality, respect-
Procedures. With the support of the administrative
ively (on average .38, .29, .32 for the three respective
staff of three middle schools from two school districts,
measures), the results support a domain- or task-specific
the divergent thinking test and the creativity-related
interpretation (Runco et al., 2006). Therefore, the four
traits questionnaire were administered to the 8th-grade
task measures were used to form a composite measure
participants in small groups. The total time offered to
of creativity. Aggregating the performance across the
participants was 50 minutes. The total time participants
four tasks not only better represents a diverse set of task
actually took to complete the tasks varied from a half
conditions set up to measure creative ideation but also
hour to 60 minutes. Participants were given either a
enhance the reliability of the measure. The alpha
movie ticket or an art craft present upon the completion
reliability for the composite measure of creativity is .86.
of the tasks as a token of thanks. Participation is volun-
Means, standard deviations, and zero-order intercor-
tary through parental consent and student assent, and
relations of all relevant variables are presented in
the reward or the token of thanks was not part of invi-
Table 1. For descriptive purposes, mean scores on each
tation (i.e., not used as incentives).
dimension for the high SES school and low SES school
are listed on the right side of the table. As can be seen,
all measures of achievement and psychosocial variables
RESULTS
are in favor of the former.
Preliminary Analysis
Effects of School and Parental Education
Since there were missing values in some cases, given the
small sample size of our study, a decision was made to To address the first research question about the differ-
impute the missing data. The missing data in ParentEd, ences in creativity between the two school districts, two
scgrade, artgrade, and AgeinMon were replaced by side-by-side boxplots were used (see Figure 1) to visualize
school means, and the missing data in the 90 Likert-type the differences. Comparing with that of the low-SES
items and in the creativity measures were imputed by school district (n ¼ 118), the creativity composite score
EM algorithm (Schafer, 1997), using SAS PROC MI. of the high-SES school district (n ¼ 111) had a wider
To reduce the dataset to manageable variables, image range and a higher mean. The two independent sample
factor analysis (Jöreskog, 1969) with varimax rotation t statistic was 6.85 with a degree freedom of 181 and a
was conducted with gender, age by month, and the 90 p-value of 0.0002, and the corresponding Cohen’s d was
self-report items, with the number of derived common 1.02.
factors determined by examining the scree plot as dis- To address the second research question regarding
cussed in Pruzek (2005). Six common factors were relative contributions of school and parent education, a
extracted, accounting for 28% of the total variance. univariate analysis of variance (ANOVA) was conduc-
The six factors were conceptually meaningful and thus ted, with the composite score of creativity as a criterion
assigned the following labels, respectively, with alpha variable, and school type (high SES vs. low SES) and
reliabilities and numbers of items reported in the parent education (two levels: undergraduate or below,
parentheses: (a) perceived teacher and parent support and graduate) as two predictors. The reason that parent
(a ¼ .77, n ¼ 12), (b) self-confidence (a ¼ .85, n ¼ 16), education only had two levels was because the high SES
(c) intrinsic cognitive motivation or need for cognition school only had four students whose parents only had
(a ¼ .84, n ¼ 14, (d) openness to experience (a ¼ .64, high school education, thus not sufficient for an
n ¼ 13), (e) adventurousness (a ¼ .74, n ¼ 10), and (f) ANOVA analysis; therefore, parent education codes
artistic imagination (a ¼ .59, n ¼ 8). Six variables were for the high school and undergraduate levels were com-
derived by averaging the item cores. Five items with bined. The results indicate that there were statistically
high complexity values and low communality values significant main effects of school type (F(1, 225) ¼ 15.12,
were not included. Other 14 items were excluded from p ¼ .0002, partial g2 ¼ .063) and parent education
SOCIAL/EDUCATIONAL ENVIRONMENTS AND CREATIVITY IN ADOLESCENCE 195
TABLE 1
Means, Standard Deviations, and Correlations of Environmental and Personal Variables Related to Creativity Capabilities and Creative
Inclinations (sample size ¼ 229)
Note. On the diagonal are alpha reliabilities. Grade in M=Sci ¼ self-reported grades in math=science; Grade in LA=SS ¼ self-reported grades in
language arts=social studies. PT Support ¼ perceived parent and teacher support for developing creativity-related characteristics; Cognitive
Motivation ¼ Need for Cognition; Openness ¼ Openness to Experience. (On the right columns).
High SES ¼ mean scores for high SES schools, Low SES school ¼ mean scores for low SES schools.
(F(1, 225) ¼ 5.6, p ¼ .019, Partial g2 ¼ .024), but the parent a graduate degree, the high-SES school district (n ¼ 21)
education by school type interaction effect was not. also outperformed the low-SES school district (n ¼ 98)
Since parent education and school type was highly cor- by a notable margin (103 vs. 82). As indicated, the stu-
related (r ¼ .64), and the main effect of school type on dents of the lower parent education background but
creativity was confounded by the effect of parental edu- who were with the high-SES school district performed
cation, the score differences between students of the two better than the students of the higher parent education
school districts with matching parent education were background but who were with the low-SES school dis-
examined. The results (Figure 2) indicate that for the trict (103 vs. 91), suggesting a school compensatory effect.
higher parental education level (parents having graduate
degrees), the high-SES school district (n ¼ 90) outper-
formed the low-SES school district (n ¼ 20) by a large Mediating Effects of Psychological Variables
margin (131 vs. 91). For those whose parents did not have To address the third research question, a hierarchical
multiple regression was computed with the composite
TABLE 2
The results of a hierarchical multiple regression analysis with creative ability (composite measure) as a dependent variable
Step 1:
School type .3 .23 .18
Parent education .19 .20
.13 .13
Step 2:
Grades combined .25 .21
Teacher support .04 .25 .06 .02
Step 3:
Confidence=Motivation .14
Adventurousness=Openness .03 .27 .015
2 2
Note. All decimals are rounded up to two digits. b is standardized regression coefficient. R D ¼R increase.
p < .05. p < .01.
SOCIAL/EDUCATIONAL ENVIRONMENTS AND CREATIVITY IN ADOLESCENCE 197
variations in school achievement, which has a direct A theoretically and practically important finding of
performance component. A distinct effect of academic the study is that creative potential, as indicated by flu-
achievement on the measure of creativity is not difficult ent, flexible, and original ideation, is associated with
to understand. The planet item and toy dog item entail two major schooling outcomes: academic achievement
creative ideation that is knowledge-rich; that is, one and intrinsic cognitive motivation. The role of intrinsic
has to have rich knowledge backgrounds to produce cognitive motivation (or need for cognition; Cacioppo
many novel ideas about what questions to ask about a & Petty, 1982), among other inclinations, provides a
planet (in the case of the planet item) and how to change bridge between the environmental conditions and cre-
structural and functional features of a toy dog. Even for ative expressions, and suggests a likely person-context
the square item, many innovative ways to divide a square reciprocal process leading to high performance on the
into four equal parts found in high performers in this measure of creativity as well as typical academic
study demonstrate good geometric knowledge as well achievement tasks.
as mathematical inventiveness, thus is to some extent As a preliminary effort to understanding the social
domain-specific (Runco et al., 2006). It is also conceiv- and educational impact on creativity, this study has its
able that even the most traditional divergent thinking limitations. School type and parent education are all
tasks such as the one used in the presented research, with social addresses (Bronfenbrenner, 1992), which are
parallel lines to draw objects, also require world knowl- social–structural but not social–psychological process
edge (Feldhusen & Goh, 1995). variables. They need to be unpacked empirically to
The third set of variables in the predictive equation, reveal their underpinnings. For example, more infor-
self-confidence and intrinsic motivation to engage in cog- mation is needed to understand how the two school dis-
nitively demanding activities (i.e., cognitive motivation), tricts differed in terms of curriculum and instruction,
was also psychosocial in nature, meant to explain and how the situation of educational resources and
long-term developmental effects on creativity, as these social capital in these neighborhoods or communities
variables show enduring individual differences and differed. Until these questions are answered, psychoso-
developmental continuity rather than varying from cial explanations for the data are still conjectural,
situation to situation. The results of the positive impact although the pattern of meditational effects that bridge
of cognitive motivation and self-confidence suggest poss- school type and parent education on the one hand and
ible mediation of both academic achievement and intrin- creative performance on the other provide important
sic motivation, arguably two of the most important clues and leads.
educational outcomes. As differences on these measures There are profound policy and practical implications
were evident between the two school districts (Table 1), it of a possible creativity gap, no less than the well-
can be inferred that school and parent effects on creativ- researched academic achievement gap. The issue not
ity may be mediated by these psychosocial variables. only concerns equity, the equal right to an enriched
At a conceptual and theoretical level, stable individ- social and educational environment, and ultimately, the
ual differences and long-term developmental changes equal opportunity for creative expression and pro-
and differences may play an important role in how ductivity; it also raises new possibility for increasing
social and educational experiences shape development social capital so that optimal adolescent development
of creativity. As indicated in the study, academic can be facilitated. Better school investment coupled with
achievement was associated with how adolescents per- parental investment in children’s intellectual develop-
formed on the creativity tasks, as well as how they were ment in terms of knowledge building and intellectual
motivated to pursue intellectually challenging tasks. interests seems to be a viable way of promoting
These variables are likely relational (reflecting a per- the development of creativity and enhancing creative
son–environment interface or ecology) rather than resid- performance.
ing either in the social context or in the person. Certain
individual temperament or personality characteristics
REFERENCES
and abilities conducive to creativity are likely meshed
with or enhanced by certain environmental structures,
Bronfenbrenner, U. (1989). Ecological systems theory. In R. Vasta
or negatively, hindered by a lack of educational (Ed.), Six theories of child development: Revised formulations and
resources and intellectually stimulating environment current issues (pp. 187–249). London: Jessica Kinsley Publishers.
(Neitzel & Stright, 2004). Nevertheless, it is unlikely that Byrnes, J. P. (2003). Factors predictive of mathematics achievement in
all the effects of parent education and school type found White, Black, and Hispanic 12th graders. Journal of Educational
in this study can be reduced to individual differences. Psychology, 95, 316–326.
Cacioppo, J. T., & Petty, R. E. (1982). The need for cognition. Journal
Thus, the findings support the argument that both indi- of Personality and Social Psychology, 42, 116–131.
vidual and environmental influences may jointly con- Cacioppo, J. T., Petty, R. E., Feinstein, J., & Jarvis, W. B. G. (1996).
tribute to creativity development (Niu, 2007). Dispositional differences in cognitive motivation: The life and times
198 DAI ET AL.
of individuals varying in need for cognition. Psychological Bulletin, Moore, M., & Russ, S. W. (2008). Follow-up of a pretend play
119, 197–253. intervention: Effects on play, creativity, and emotional processes
Costa, P. T., Jr., & McCrae, R. R. (1992). Revised NEO personality in children. Creativity Research Journal, 20, 427–436.
inventory (NEO PI-RTM) and NEO five-factor inventory Mumford, M. D., Marks, M. A., Connelly, M. S. Zaccaro, S. J.,
(NEO-FFI): Professional manual. Odessa, FL: Psychological Johnson, J. F. (1998). Domain-based scoring of divergent-thinking
Assessment Resources. tests: Validation evidence in an occupational sample. Creativity
Cramond, B., Matthews-Morgan, J., Bandalos, D., & Zuo, L. (2005). Research Journal, 11, 151–163.
A report on the 40-year follow-up of the Torrance Tests of Creative Murdock, T. B. (2000). Incorporating economic context into edu-
Thinking: Alive and well in the new millenium. Gifted Child cational psychology: Methodological and conceptual challenges.
Quarterly, 49, 283–291. Educational Psychologist, 35, 113–124.
Cropley, A. J. (2000). Defining and measuring creativity: Are creativity Neitzel, C., & Stright, A. D. (2004). Parenting behaviours during child
tests worth using? Roeper Review, 23, 72–79. problem solving: The roles of child temperament, mother education
Dai, D. Y., & Shen, J.-L. (2008). Cultivating creative potential during and personality, and the problem-solving context. International
adolescence: A developmental and educational perspective. Korean Journal of Behavioral Development, 28, 166–179.
Journal of Thinking and Problem Solving, 18, 83–92. Niu, W. (2007). Individual and environmental influences on Chinese
Feldhusen, J. F., & Goh, B. E. (1995). Assessing and accessing creativ- student creativity. Journal of Creative Behavior, 41, 151–175.
ity: An integrative review of theory, research, and development. Ogletree, E. (1971). A cross-cultural examination of the creative think-
Creativity Research Journal, 18, 231–247. ing ability of public and private school pupils in England, Scotland,
Frederickson, N., & Petrides, K. V. (2008). Ethnic, gender, and and Germany. Journal of Social Psychology, 83, 301–302.
socio-economic group differences in academic performance and sec- Plucker, J. A. (1999). Is the proof in the pudding? Reanalyses of
ondary school selection: A longitudinal analysis. Learning and Indi- Torrance’s (1958 to present) longitudinal data. Creativity Research
vidual Differences, 18, 144–151. Journal, 12, 103–114.
Gottfried, A. W., Gottfried, A. E., Bathurst, K., & Guerin, D. W. Prabhu, V., Sutton, C., & Sauser, W. (2008). Creativity and certain
(1994). Gifted IQ: Early developmental aspects: The Fullerton longi- personality traits: Understanding the mediating effect of intrinsic
tudinal study. New York, NY: Plenum. motivation. Creativity Research Journal, 20, 53–66.
Gough, H. G. (1987). California Psychological Inventory administra- Pruzek, R. M. (2005). Factor analysis: Exploratory. In B. S. Everitt &
tor’s guide. Palo Alto, CA: Consulting Psychologists Press. D. C. Howell (Eds.), Encyclopedia of statistics in behavioral science
Grissmer, D. W., Williamson, S., Kirby, S. N., & Berends, M. (1998). (vol. 2, pp. 606–617). Chichester, UK: John Wiley & Sons.
Exploring the rapid rise in Black achievement scores in the United Rothenberg, A. (1979). The emerging goddess. Chicago, IL: University
States (1970–1990). In U. Neisser (Ed.), The rising curve: Long-term of Chicago Press.
gains in IQ and related measures (pp. 251–285). Washington, DC: Runco, M. A., Dow, G., & Smith, W. R. (2006). Information, experi-
American Psychological Association. ence, and divergent thinking: An empirical test. Creativity Research
Harrington, D. M., Block, J. H., & Block, J. (1987). Testing aspects of Journal, 18, 269–277.
Carl Rogers’s theory of creative environments: Child-rearing Schafer, J. L. (1997). Analysis of incomplete multivariate data. London,
antecedents of creative potential in young adolescents. Journal of UK: Chapman & Hall.
Personality and Social Psychology, 52, 851–856. Shen, J.-L. (2005). The creativity test [Unpublished]. Beijing, China:
Jöreskog, K. G. (1969). A general approach to confirmatory maximum Normal University.
likelihood factor analysis. Psychometrika, 34, 183–202. Torrance, E. P. (1972). Predictive validity of the Torrance Tests of
Lichtenwalner, J. S., & Maxwell, J. W. (1969). The relationship of Creative Thinking. Journal of Creative Behavior, 6, 236–252.
birth order and socioeconomic status to the creativity of preschool Tsao, Y. (2008). Using guided play to enhance children’s conversation,
children. Child Development, 40, 1241–1247. creativity and competence in literacy. Education, 128, 515–520.
Liu, Y. (1970). Cattell 16 PF administrator’s manual. Beijing, China: Vijayalakshmi, J. (1980). Academic achievement and socio-economic
Beijing Normal University. status as predictors of creative talent. Journal of Psychological
McLoyd, V. C. (1998). Socioeconomic disadvantage and child devel- Researches, 24, 43–47.
opment. American Psychologist, 53, 185–204. Wang, X. (2003). A study about students’ creative tendency and their
Milgram, R. M. (1983). Validation of ideational fluency measures of perception of teachers’ classroom behavior [Unpublished Master’s
original thinking in children. Journal of Educational Psychology, thesis]. Beijing, China: Normal University.
75, 619–624. Williams, F. E. (1980). Creativity assessment packet. Buffalo, NY: D.
Miller, B. C., & Gerard, D. (1979). Family influences on the develop- O. K Publishers.
ment of creativity in children: An integrative review. Family Coordi- Zhang, Z. (2005). Software for divergent thinking tests [Unpublished].
nator, 28, 295–312. Beijing, China: Normal University.
SOCIAL/EDUCATIONAL ENVIRONMENTS AND CREATIVITY IN ADOLESCENCE 199
APPENDIX A. CODES AND CATEGORIES spatial usage, things for building, things for physical
OF DIVERGENT THINKING AND culture, musical instruments, toy, dresses, sight, weapon,
COMPUTERIZED SORTING ghosts and spirits, abstractions, and experimental
equipment
14. Categories for the Planet Question Computer Program (Zhang, 2005) for data sorting
and management: Screen display (left column: words;
ET, biology, resource, structural properties, climate,
right column: categories)
physical environment, physical attributes, time, space
relationship, origins and future, survival, science=
technology=military, exploiture, and special sight