Thinking Skills and Creativity 24 (2017) 250–267

Contents lists available at ScienceDirect

Thinking Skills and Creativity

journal homepage: www.elsevier.com/locate/tsc

Exploring creative learning in the classroom: A multi-method MARK

approach☆
⁎
Aleksandra Gajdaa, Ronald A. Beghettob, Maciej Karwowskia,
a
Department of Educational Sciences, The Maria Grzegorzewska University, Szczesliwicka St., 40, 02353, Warsaw, Poland
b
Department of Educational Psychology, Neag School of Education, University of Connecticut, Storrs, CT, USA

AR TI CLE I NF O AB S T R A CT

Keywords: How might researchers better understand the variations in creative learning in and across
Mix methods classrooms? This article addresses this question. We introduce a multi-method approach that we
Creative learning used to explore the more dynamic features of creative learning in ten elementary classrooms. The
Classrooms ten classrooms were first classified into one of three groups (positive, negative, and null), based
Creativity
on the relationship between students’ (N = 204) measured creativity and academic achievement
Academic achievement
Teachers
(average positive r = 0.52; average negative r = −0.23; and average null r = 0.02). Next, we
Students analyzed observed teacher and student behaviors in each classroom. We found different patterns
of behavior based on classroom classifications. Teachers in classrooms with a positive association
between creativity and academic achievement tended to demonstrate more caring behaviors
toward students and to provide more emotional support to students. We also found that teacher
behaviors associated with encouraging creativity in the classroom were associated with students’
positive engagement, self-expression, and ideation (regardless of classroom type). Finally, we
used a micro-level interactional analysis to visually illustrate patterns of interactions between
teachers and students in three different classroom classifications. We found more extended and
exploratory interactions in the positive association classroom, whereas the negative association
classroom was characterized by more directive and rapidly closing patterns of interaction and the
null association classroom tended to have patterns of interaction that left students’ ideas
suspended and lacked exploration, development or refinement of ideas. We close by discussing
how the use of blended methodologies, like the one demonstrated in this article, can be further
refined and developed in subsequent research to explore and understand the more dynamic
features of creative thought and action in classroom settings.

“The problems of creativity in the educational setting are endless, and the scope of research in this area is rapidly spreading” Guilford
(1967, p. 10)
How might creativity researchers better understand the relationship between creativity and learning in the classroom? Typically,
researchers have attempted to examine this relationship by exploring patterns of association amongst measures of creativity and
academic achievement (e.g., Gajda, 2016; Freund & Holling, 2008). The largest meta-analysis, to date, which has examined this
association (Gajda, Karwowski, & Beghetto, 2017) indicates a positive, albeit moderate, association between creativity and academic
achievement (r = 0.22). However, both the direction and strength of the relations between creativity and academic achievement is
highly variable across schools and grades (Freund & Holling, 2008). Indeed, previous studies have demonstrated seemingly

☆
The preparation of this article was supported by grant from the National Science Centre, Poland (UMO-2011/03/N/HS6/05073).
⁎
Corresponding author.
E-mail address: [email protected] (M. Karwowski).

http://dx.doi.org/10.1016/j.tsc.2017.04.002
Received 24 September 2016; Received in revised form 6 February 2017; Accepted 10 April 2017
Available online 22 April 2017
1871-1871/ © 2017 Elsevier Ltd. All rights reserved.
A. Gajda et al. Thinking Skills and Creativity 24 (2017) 250–267

contradictory results. Gralewski and Karwowski (2012) have, for instance, found that in some classrooms correlations between
learning and achievement were positive and strong, whereas in other classrooms the correlation was non-existent and, in some cases,
negative. Although previous work exploring the association between measures of creativity and academic learning highlights how the
relationship between creativity and academic learning differs across classrooms contexts, such studies fall short in capturing the more
dynamic, socio-psychological processes that may help explain these variations across classrooms.
What is needed are blended methodologies that can simultaneously examine the more dynamic, qualitative, and micro-level
interactions amongst teachers and students in light of the more static, quantitative, and larger grain differences in associations
between creativity and academic learning across classrooms. Although there have been some initial efforts in this regard (see
Beghetto, 2016a; Tanggaard & Beghetto, 2015; Schacter, Thum, & Zifkin, 2006), we would argue that more integrated and
comprehensive approaches can yield new and needed insights into the relationship between creativity and academic learning.
The purpose of the paper is to demonstrate a blended approach that we used to explore the more dynamic and micro-level features of
student and teacher behaviors in classrooms characterized by larger grain differences in creativity and academic achievement.
Specifically, we report on our initial efforts aimed at blending different analytic approaches in an effort to explore whether and how
teacher and student interactions differ across elementary classrooms that have three different patterns of association between
creativity and learning: positive, null, and negative.

1. Creativity and learning

Researchers have long recognized the link between creativity and learning (Dewey, 1899/2007; Dewey, 1934/2005; Guilford,
1950; MacClelland, 1975; Ripple, 1977). Although researchers agree that learning and creativity are related, they tend to highlight
different aspects of this relationship. Some, for instance, have asserted that learning represents a special case of the creative process
(Starko, 1995), whereas others have asserted that creativity represents an instance of learning (Guilford, 1950). In fact, some have
gone as far as to assert that creativity and learning essentially represent the same phenomenon (Barron, 1967). Most, however,
recognize that creativity and learning have a mutually supportive relationship with each other (Beghetto, 2016b).
A key question, at this point, is what kinds of teacher and student behaviors are supportive of creative learning? In the section that
follows, we briefly review previous theories and research that highlights features of the learning environment that are thought to be
supportive of creative learning.

1.1. The role of teachers

Creative learning requires opportunities for students to engage with new and different perspectives and have opportunities to
share and receive feedback on their own unique perspectives (Beghetto, 2016b; Glăveanu & Beghetto, 2016). The result of doing so
can simultaneously support the development of new and personally meaningful learning for students’ and provide opportunities for
students to contribute to their peers’ and teachers’ learning (Beghetto, 2016b, 2016c). Not surprisingly, then, the kinds of interactions
and behaviors demonstrated and expected by teachers play a key role in determining whether creative learning will be supported or
suppressed (Beghetto, 2016c; Feldhusen & Treffinger, 1980; Soh, 2000).
Supportive teacher behaviors outlined by creativity researchers (Beghetto, 2013; Cropley, 2001; Davies et al., 2012; Hong,
Hartzell, & Greene, 2009; Schacter et al., 2006; Torrance, 1962) include:

• establishing improvement-focused learning goals;

• providing opportunities for students to use their imagination while learning;
• encouraging students to take sensible risks and act independently;
• teaching with a more game-like or playful approach;
• providing opportunities for choice and discovery;
• encouraging flexible thinking and confidence in students ideas;
• treating student questions and ideas (especially unusual and unexpected ones) seriously;
• refraining from premature assessment of students ideas;
• demonstrating a belief that teaching should go beyond imparting simplistic and factual knowledge; and
• supporting students when they fail by showing them ways to learn from their mistakes.
Implementing these strategies, however, seems to be easier said than done. Indeed, teachers who espouse an appreciation and
acceptance of student creativity may not demonstrate those beliefs in practice (Chan & Chan, 1999; Esquivel, 1995;
Gralewski & Karwowski, 2013, 2016; Scott, 1999; Westby & Dawson, 1995). Consequently, instructional practices conducive to
creativity are not always easy to find in observations of classroom teaching – particularly in classrooms with high proportions of
minority and low-performing students (Schacter et al., 2006).
Conversely, the potentially stifling practice of asking known-answer questions and evaluating students for how well they can
reproduce what is expected and how it is expected may be more readily observed (Goodlad, 2004; Mehan, 1979). We would assert
that this likely has less to do with teachers harboring a secret dislike for creativity or creative students and more likely has to do with
teachers not knowing how to simultaneously support creativity and learning and, instead, feel that they must choose supporting one
over the other (Beghetto, 2007). One reason this occurs is because teachers (like most people) tend to narrowly define creativity
(Andiliou & Murphy, 2010; Gralewski & Karwowski, 2016) – associating it with other personality (Scott, 1999) or cognitive traits

251
A. Gajda et al. Thinking Skills and Creativity 24 (2017) 250–267

(Karwowski, 2007), which may be at odds with the goals and objectives of classroom teaching and learning.

1.2. The role of students

Students also play a key role in establishing a learning environment supportive of creative learning. Not only do students need to
have the confidence and willingness to take the risk necessary to share their unique perspectives and ideas, they also need to be
willing to listen to and support their peers in doing the same (Beghetto, 2016b). This includes knowing how to build on each other’s
ideas and productively contribute to the learning task or activity. Indeed, even teachers who genuinely appreciate creative learning
may view students’ unique perspectives and ideas as disruptive when they start to drift off-topic and no longer connect to the task at
hand (Beghetto & Plucker, 2015).
Given that students are still in the process of developing their creative capacity, it makes sense that it may be difficult for them to
regulate how and when they express their creativity. This may be particularly challenging for students who are characterized as
“rebellious creative” (Karwowski, 2017; Karwowski & Jankowska, 2016), who tend to demonstrate non-conformist and independent
behaviors and thereby do not easily fit into the sometimes rigid framework imposed upon them by educational institutions
(Günçer & Oral, 1993; Westby & Dawson, 1995).
The variability of student creative expression can pose challenges for both students and teachers. Johnson and Hatch (1990), for
instance, reported on case studies of creative children – who were, at turns, able to be highly engaged and demonstrate on-task
behavior and other times became disruptive. One child (Jack) became disruptive in class when he was not able to persuade his peers
to his way of thinking or when he did not feel like engaging in an activity. The teacher explained to the researchers that Jack’s
creativity could sometimes become disruptive when he wanted “to try his own ideas, to move on, to advance forth … not sit and listen
to someone else’s opinion” (p. 216). These descriptive observations highlight how students’ creative behaviors can vary between and
within children and, in some cases, are at odds with teachers’ behavioral expectations.
Taken together, the behaviors and interactions demonstrated amongst teachers and students help establish learning environments
that are more or less supportive of creative learning. If students feel that unexpected questions and answers are not encouraged and
respected, then they likely will not be willing to take the risks necessary to share or explore new ideas (Kawenski, 1991) and, instead,
take the safer path of trying to provide the answer that there teachers and peers expect rather than experience being dismissed, down-
graded, or even ridiculed (Beghetto, 2013, 2016c).
In order to better understand how creative learning varies in and across classroom environments, creativity researchers need an
analytic approach that allows for exploring the behaviors and interactions of teachers and students in classrooms that demonstrate
varying patterns of creative learning. This is the goal of the present study.

2. The present study

The aim of this study is to illustrate an integrative analytic approach that combines quantitative classifications of classrooms
based on correlational analyses of creativity and academic learning with a more micro-level, qualitative exploration of specific
behaviors and interactions of teachers and students in and across classrooms. With respect to the quantitative classifications, we focus
on identifying three types of classrooms: those that demonstrate a positive relationship between student creativity and academic
learning (positive); those that demonstrate a negative relationship (negative); and those that demonstrate no relationship between
creativity and academic learning (null).
We then examine patterns of observed teacher and student behaviors in those classrooms and, finally, zoom-in even closer and
highlight more micro-level patterns of interactions amongst teachers and students engaged in classroom discussions. We focus our
attention on analyzing classroom discussions because they represent one of the most frequently used instructional strategies by
teachers (Cazden, 2001) and because discussions also provide students with opportunities to share and test out their unique ideas and
perspectives on the subject matter being taught (Beghetto, 2016a, 2016b).

3. Method

3.1. Selection of classrooms and participants

We used purposive sampling to select 10 elementary classrooms, 204 elementary students (107 girls and 97 boys), and 10 teachers
(all females) for this study. More specifically, we drew our sample of classrooms and participants from a large cross-sectional dataset
(Gajda, 2016), which included more than 1000 students and 20 elementary classrooms. Using correlational analysis, we selected 10
elementary classes that demonstrated different patterns of relationship between students’ creative ability and school grades (see
below for a discussion of these measures). The 10 classrooms selected for this study represented the following categories: positive
relationship between student creativity and academic achievement (n = 3, average r = 0.52, range from r = 0.44, to r = 0.51 and
r = 0.60), negative relationship (n = 3, average r = −0.23, range from r = −0.34, to r = −0.25 and r = −0.10) and null
relationship (n = 4, average r = 0.02, range from r = −0.02, to r = 0.03 (two classes) and r = 0.04).
The 10 classes selected for this study had an average class size of 20 students (min = 15, max = 28) and represented five schools
that were similar in terms of localization, average socioeconomic status, and academic school performance. Consent for students’
participation in the study was obtained in writing from the students’ parents. In cases where parents or guardians had any questions
pertaining to the purpose of the study, the first author provided explanatory information. Six parents had not consented to their

252
A. Gajda et al. Thinking Skills and Creativity 24 (2017) 250–267

child’s participation in the creativity test, while all parents gave consent to observational study.

3.2. Measures and procedure

We used four types of data-sources for this study: measures of student creativity, measures of student academic achievement,
observational measures of teacher and student behavior, and audio-recorded teacher and student interactions.

3.2.1. Student creativity

We used Urban and Jellen’s Test of Creative Thinking – Drawing Production (TCT-DP) (Jellen & Urban, 1986; Urban & Jellen,
1996) to measure creative ability. The test’s sheet included six incomplete shapes, which each participant used as the basis for
completing a drawing in his or her own way. Each participant was allotted 20 min to complete the task.
The final drawing of each student was then quantitatively and qualitatively assessed using 14 criteria, which were used to
calculate a composite creativity score (Urban & Jellen, 1996): Continuation (Cn), Completions (Cm), New elements (Ne), Connections
made with a line (Cl), Connections made to produce a theme (Cth), Boundary breaking (fragment-dependent) (Bfd), Boundary
breaking (fragment-independent) (Bfi), Perspective (Pe), Humor and affectivity (Hu), Unconventionality: manipulation of the test
material (Cca), Unconventionality: surreal or abstract drawings (Ccb), Unconventionality: use of symbols or signs (Ccc),
Unconventionality: Non-stereotypical use of a certain element (Ccd) and Speed (Sp). The overall reliability was α = 0.71, which
is in accordance with levels obtained in previous studies (e.g., Karwowski & Gralewski, 2013).

3.2.2. Academic achievement

We used students’ average school grades (GPA)1 from the previous semester as a proxy of academic achievement (M = 4.40,
SD = 0.91, range from: 1 = low to 6 = excellent). There were no statistically significant, F(2212) = 0.68; p = 0.512 differences in
GPA across classes (positive correlation classes: M = 4.35, SD = 1.14, N = 65, min = 1, max = 6; negative correlation classes
M = 4.33, SD = 0.77, N = 64, min = 2, max = 6, null correlation classes M = 4.51, SD = 1.14, N = 84, min = 1, max = 6,
SD = 1.14).

3.2.3. Teacher and student behavior

The first author conducted observations in each of 10 classes. Observed classes included five Polish (native) language class and
five math classes. During the observations, the researcher documented teacher and student behaviors and interactions using an
observation protocol developed for this study (see Appendices A and B for examples of behaviors coded). The observation protocol
consisted of a number of students’ and teachers’ exemplary behaviors, listed with regard to three components of creativity as defined
in the Typological Model of Creativity (Karwowski, 2017; Karwowski and Jankowska, 2017): creative ability, openness and
independence. The behaviors were dichotomously coded (0 = behavior absent, 1 = behavior present,) in five-minute intervals, which
resulted in a total of 90 observational units, clustered within 10 classes for teachers’ and students’ behaviors.

3.2.4. Interactions between teachers and students

The first author audio-recorded classroom interactions between teachers and students in 10 classrooms. These data were collected
for the purpose of conducting a micro-level interactional analysis (Beghetto, 2016a; Tanggaard & Beghetto, 2015), which uses
transcribed excerpts of those interactions to develop visual diagrams of the patterns of interaction amongst teachers and students that
can then be used for further analysis and interpretation. Doing so provides researchers with an analytic artifact (beyond the
transcript) that can be helpful in identifying patterns of interactions, which may be more or less supportive of creative learning (see
Beghetto, 2016a, 2016b; Tanggaard & Beghetto, 2015).
In the present study, we used this approach to visually illustrate and explore patterns of interaction amongst teachers and students
engaged in academic discussions in three different classrooms: one classified as having a null relationship, one with a negative
relationship, and one with a positive relationship between creativity and academic achievement. We used the following criteria to
select a sample excerpt to apply this approach. First, a teacher and at least two students had to be engaged in an academic discussion
(i.e., talking about academic subject matter, rather than some other procedural or non-instructional exchange). Second, the
discussion had to include an exchange in which questions were posed and opportunities for students to respond were made available.
Finally, the discussion had to be self-contained (i.e., an interaction focused on a particular academic goal) and the audio had to be
clear enough throughout the discussion to transcribe. Using these selection criteria, we were able to select three relatively brief
excerpts of teacher-student interactions (i.e., ranging from 13 to 29 total utterances) representing each classroom type (negative, null,
and positive).

1
A recent meta-analysis (Gajda et al., 2017) demonstrates that the relationship between school grades and creativity is weaker than a relationship between
creativity and achievement tests results. Given that we did not have complete standardized achievement data to use in this study, we relied on average school grades
(which is common in such studies, see Gralewski & Karwowski, 2012). Although our achievement data does not seem to be limited by restricted variance (i.e., very
high or very low grades), it is possible that the potentially lower reliability of teacher grades (as compared to standardized assessments) could result in artificially
under-estimated relationships between academic achievement and creativity. In the case of this study, under-estimates of the relationship between creativity and
academic achievement likely would not make a substantive difference in how the 10 classes were classified. Still, we want to highlight this potential limitation.
2
The number of degrees of freedom for denominator is higher than the number of participants in observational study, as we used data about GPA of all students
(also these who did not solve creativity test), to obtain unbiased estimate of classroom average school achievement.

253
A. Gajda et al. Thinking Skills and Creativity 24 (2017) 250–267

The first author prepared a transcript of each brief interaction and then we used the transcript as the basis for depicting the
interaction using a visual diagram3 (adapted from Tanggaard & Beghetto, 2015; see also Beghetto, 2016a). More specifically, we
plotted teacher and student utterances along a horizontal timeline. We then traced out patterns of interactions as they moved in and
between more or less open (i.e., indeterminate and exploratory) or closed (i.e., determinate or fixed) horizons of the diagram. We also
added notation to differentiate between different speakers (e.g., teacher, students), different utterances from the same speakers; and
moments in the interactions when ideas were accepted, contested, dismissed, suspended, and returned to in a subsequent interaction.

4. Results

4.1. Analysis of teacher and student behaviors

We used exploratory factor analysis to examine the factor structure of the observed teacher and student behaviors. More
specifically, we used exploratory factor analysis in the structural equation modeling (ESEM) scheme, which treated individual
behaviors as measured on the count scale and used the robust weighted least squares estimator (WLSMV, Nussbeck, Eid, & Lischetzke,
2006). We conducted the analyses using Mplus software (Muthén & Muthén, 1998–2014) and controlled for students’ class grouping.

4.1.1. Teacher behaviors

With respect to teacher behaviors, our results indicate that a four-factor solution fit the data well (Hu & Bentler, 1999; Kline,
2011): CFI = 0.97, TLI = 0.96, RMSEA = 0.023 (90% CI: 0.000, 0.055), WRMR = 0.62 (see Table 1).
As displayed in Table 1, the first factor represents behaviors associated with caring for students (e.g., careful listening, authenticity
of the teacher’s behaviors, delayed assessment of ideas). The second factor represented behaviors associated with encouraging
creativity, (e.g., encouraging students to be inventive and asking open questions). The third factor represents behaviors associated
with emotional support (e.g., friendly humor, motivating to manage failure). The fourth factor represented risk acceptance behaviors
(e.g., testing new ideas, taking risks, accepting unconventional ideas).
In an effort to explore the most commonly observed behaviors amongst teachers, we calculated factor scores.4 Given that teachers’
behaviors were coded as either present (code = 1) or absent (code = 0), we first estimated internal consistency of four factors
obtained in factor analysis. To this end we used Cronbach’s alpha estimated on a matrix of tetrachoric correlations, which indicated
acceptable internal consistency of our factors: caring (α = 0.79), encouraging creativity (α = 0.91), emotional support (α = 0.81),
and risk acceptance (α = 0.75). In the next step, we proceeded with repeated-measures ANOVA to explore potential differences
amongst behaviors. Our result indicate significant differences amongst these behaviors, with caring behaviors being the most typically
observed set of teacher behaviors (M = 0.32, SD = 0.35), followed by encouraging creativity (M = 0.16, SD = 0.26), providing
emotional support (M = 0.11, SD = 0.19) and, finally risk acceptance (M = 0.06, SD = 0.10), F(3, 89) = 20.33, p < 0.001,
η2 = 0.19.

4.1.2. Student behaviors

With respect to student behaviors, a four-factor solution fit the data well: CFI = 0.94, TLI = 0.89, RMSEA = 0.039 (90% CI: 0.00,
0.08), WRMR = 0.70 (see Table 2).
The first factor represented student misbehavior (i.e., behaviors that challenged the established order and classroom rules, such as
questioning classroom rules and providing unexpected or inappropriate responses). The next factor represented positive engagement
(i.e., not being discouraged by setbacks and demonstrating curiosity and interest in the topic). The third factor included behaviors
that represented self-expression (i.e., expression of one’s own views, making jokes and using humor). Finally, the fourth factor
represented behaviors associated with ideation (i.e., generating many ideas from various categories, providing surprising answers).
As with teacher behaviors, we estimated the internal consistency of our factors. Similar to what we found with the factors of
teacher behaviors, the reliability of student factors was acceptable: misbehavior (α = 0.68), positive engagement (α = 0.61), self-
expression (α = 0.79), and ideation (α = 0.77). Next, we used repeated-measures ANOVA to examine potential differences amongst
behaviors. Our results indicate that positive engagement was the most typically observed set of student behaviors (M = 0.26,
SD = 0.23), followed by ideation (M = 0.14, SD = 0.20), misbehavior (M = 0.12, SD = 0.18) and self-expression (M = 0.11,
SD = 0.18), F(3, 89) = 11.93, p < 0.001, η2 = 0.12.
Our next step was to examine whether and how these behaviors differed across classrooms that were classified as having a
positive, negative, and null relationships between creativity and academic achievement.

3
The development of micro-level interactions diagrams allows us to retrospectively depict a sample of utterances between students and teachers. Given that this
type of analysis is not based on a representative sample of all behaviors in or across selected classrooms, it should be regarded as idiographic illustration of lesson
discussion. Still, the value of such an approach is that allows researchers to depict – in a very concrete way – the more dynamic and somewhat tacit patterns of
interaction that otherwise might go unnoticed using traditional transcript or discourse analysis. This approach provides researchers with an additional artifact (beyond
the transcript) that can be used for analysis and interpretation.
4
The factor scores represent averages of the dichotomous items that make up the factor. Therefore, the means can be interpreted as a percentage of behaviors typical
for each factor.

254
A. Gajda et al. Thinking Skills and Creativity 24 (2017) 250–267

Table 1
Factor analysis of observed teacher behaviors.

Behavior unit – Teachers % Caring Encouraging Creativity Emotional Support Risk Acceptance

Careful listening to questions 13% 0.95

Delayed assessment of ideas 8% 0.85
Group work organization 20% 0.69
Authenticity, genuineness 16% 0.60
Encouraging to search for diverse ideas 19% 0.96
Encouraging to create many ideas 24% 0.90
Open questions 20% 0.90
Encouraging divergent thinking 18% 0.88
Atmosphere of creative play 23% 0.61
Friendly humor 27% 0.90
Using constructive criticism 8% 0.67
Motivating to manage failures 9% 0.48
Sensitivity to feelings 3% 0.47
Making it possible to choose and make decisions 6% 0.59
Acceptance of unconventional and strange ideas 6% 0.59
Encouraging to test new ideas 4% 0.49
Tolerance of otherness and new ideas 6% 0.45
Encouraging to imagine different things 4% 0.37

Note. N = 90 observations; % = percent amongst all behaviors observed. Loadings are estimated using the WLMSV estimator and geomin rotation. Only statistically
significant loadings are presented.

Table 2
Factor analysis of observed student behaviors.

Behavior unit – Students % Misbehavior Positive Engagement Self-Expression Ideation

Questioning classroom rules 6% 0.93

Behavioral risk 16% 0.46
Distraction of attention 22% 0.46
Inappropriate answers 3% 0.45
Transforming the student’s own and other students’ ideas 4% 0.70
Expressing emotion 3% 0.66
Curiosity and interest in the topic 44% 0.59
Opposing the teacher’s views 4% 0.44
Engagement in new tasks 43% −0.71
Ease in recalling words and expressions 11% 0.87
Expressing one’s own views 13% 0.84
Jokes and sense of humor 12% 0.32
Many ideas from various categories 10% 0.98
Many answers to the same question 26% 0.53
Surprising answers 10% 0.67

Note. N = 90 observational units; % = percent amongst all behaviors observed (percentages do not add up to 100%, as different behaviors might occur
simultaneously). Loadings are estimated using the WLMSV estimator and geomin rotation. Only statistically significant loadings are presented.

4.2. Exploring variations in teacher and student behaviors

In an effort to examine whether teacher and student behaviors varied between classroom categories, we tested two separate
multilevel models – one for teacher behaviors and one for student behaviors. Each time, we used class type to predict the four
behavioral factors. We used the null-relationship class as the reference group for our models. We also included a “lesson duration”
variable in our models, which represented nine 5-min time intervals across the entire 45-min lesson.
Table 3 presents a summary of coefficients and significance levels obtained in the teacher and student multi-level models.

4.2.1. Teacher behaviors

As displayed in Table 3, there was a significant difference in teachers’ caring behaviors between class types. Specifically, teacher’s
caring behaviors were more frequently demonstrated in classes classified as having a positive correlation between students’ creativity
and academic achievement as compared to classes with null correlation.
With respect to teacher behaviors associated with encouraging creativity, there were no significant differences between class
types. A marginal significant difference (i.e., p < 0.10) in encouraging creativity, however, was found across lesson duration in all
classroom types. Specifically, teachers’ behaviors associated with encouraging creativity decreased across the duration of the 45-min
lesson.
With respect to teachers providing emotional support to students, our results indicate significant differences between class type
and lesson duration. More specifically, teacher behaviors associated with emotional support were significantly less frequent in classes

255
A. Gajda et al. Thinking Skills and Creativity 24 (2017) 250–267

Table 3
Class Type and Lesson Duration as Predictors of Observed Teachers’ and Students’ Behaviors.

Predictor DV: Teachers’ Behaviors

Caring Encouraging Creativity Emotional Support Risk Acceptance

Class Type – Positive *

0.36 (0.14) −0.21 (0.14) 0.22 (0.18) −0.03 (0.11)
Class Type – Negative 0.02 (0.11) −0.19 (0.11) −0.17* (0.07) −0.03 (0.13)
Lesson Duration −0.02 (0.09) −0.15# (0.08) −0.20** (0.06) 0.05 (0.10)

Predictor DV: Students’ Behaviors

Misbehavior Positive Engagement Self-Expression Ideation

Class Type – Positive −0.16 (0.11) −0.12 (0.14) −0.10 (0.20) −0.20 (0.13)
Class Type – Negative −0.23* (0.08) −0.21* (0.09) −0.20* (0.09) −0.19 (0.12)
Lesson Duration 0.12 (0.14) −0.34*** (0.08) 0.14 (0.08) 0.01 (0.09)

Note. All coefficients are standardized, robust standard errors in parentheses. Class Type-Positive = classes with positive correlation creative ability-school
achievement, Class Type-Negative = classes with negative correlation creative ability-achievement (classes with null correlations served as the reference category).
#
p < 0.10.
* p < 0.05.
** p < 0.01.
*** p < 0.001.

classified as having a negative correlation between creativity and achievement (as compared to classes with the null correlation).
Similar to encouraging creativity, however, the frequency of teachers demonstrating emotional support decreased as the lesson
progressed (regardless of class type). Finally, teachers demonstrating risk acceptance behavior did not differ between class types or
across lessons duration.

4.2.2. Student behaviors

As reported in Table 3, our results indicate that student misbehavior, positive engagement, and self-expression tended to be lower
in classrooms classified as having a negative relationship between creativity and achievement (as compared to the null classroom).
Our results also indicate that across all class types, student behaviors associated with positive engagement tended to weaken in the
final fifteen minutes of the lesson.

4.3. Relationship between teacher and student behaviors

Next, we explored the relationship between observed teacher and student behaviors. Given the ordinal nature of our observational
behavior scales5, we used polychoric correlations to examine this relationship. As displayed in Table 4, there were significant
relationships (i.e., p < 0.05) between observed teacher and student behaviors. Specifically, teachers’ caring behaviors were
positively related to students’ positive engagement (r = 0.25) and ideation (r = 0.38). Moreover, teachers’ encouraging creativity
behaviors were associated with students’ positive engagement (r = 0.26), self-expression (r = 0.47), and ideation (r = 0.39). Finally,
teachers’ risk acceptance behaviors were related to students’ self-expression (r = 0.28) and ideation (r = 0.37). We found no
significant relationship between teachers’ emotional support behaviors and any observed student behaviors.

4.4. Analysis of micro-level interactions

Finally, we developed visual displays of interactions (Tanggaard & Beghetto, 2015) to help illustrate and explore patterns of
interaction between teachers and students in three different classroom types (i.e., classified as having a null, negative, or positive
relationship between creativity and academic achievement). In what follows, we provide a transcript of the class discussion, a visual
display of that discussion, and a brief analysis of the patterns of interaction for each of the three classroom types: Null, negative, and
positive.

4.4.1. Teacher-student interaction in a null classroom

The following transcript of a teacher-student interaction is from a classroom classified as having a null relationship between
creativity and academic achievement (r = 0.02; n = 20). The interaction occurs as part of a 4th grade Polish language lesson.
Teacher (U1T) (…) let’s imagine a different use for a comb than just combing your hair
(…)

5
Given observed behaviors were coded as either not present (0) or present (1), we treat our factors (composed of these scores) as measured on ordinal or even count
scale. We therefore used polychoric correlations, controlling for nesting observation duration within lessons. Readers interesting in raw Pearson’s correlations are
encouraged to contact the corresponding author.

256
A. Gajda et al. Thinking Skills and Creativity 24 (2017) 250–267

Table 4
Polychoric correlations between teacher and student behaviors.

T1 T2 T3 T4 S1 S2 S3 S4

T1: Caring 1
T2: Encouraging creativity −0.08 1
T3: Emotional support 0.16 0.15 1
T4: Risk acceptance 0.21 0.10 0.39 1
S1: Misbehavior −0.06 0.09 −0.04 0.12 1
S2: Positive engagement 0.25 0.26 0.15 0.05 −0.33 1
S3: Self-expression 0.11 0.47 0.07 0.28 0.58 0.08 1
S4: Ideation 0.38 0.39 0.12 0.37 −0.05 0.63 0.45 1

Note. Bolded coefficients are statistically significant (p < 0.05), N = 90.

Student (U2S1) Yy you can scratch your back …

Student (U3S2) I wanted to say that!
Student (U4S3) So did I!
Teacher (U5T) Wait, stop. A tool for scratching, tadek [boy’s name, transl.]
Student (U6S3) It may also serve as something to digging out something else
Student (U7S4) Indeed, I personally do that sometimes.
[Noise]
Teacher (U8T) Kamila? [girl’s name, transl.]
Student (U9S5) You can play on it
[Noise]
Student (U10S6) Play on it?
Teacher (U11T): Karolina? [girl’s name, transl.]
Student (U12S6) Yy you can attach it to your hair
Teacher (U13T) Other than combing, attaching to your hair?
Student (U14S7) a decoration
Teacher (U15T) Well, OK
Student (U16S8) a decoration?
Student (U17S6) And how can you turn it into an instrument?
Teacher (U18T) Kacper [boy’s name, transl.]
Student (U19S9) A rake!
Teacher (U20T) Weronika? [girl’s name, transl.]
Student (U21S10) You can use it as a shish-kebab stick
Student (U22S11) but how?
[shouting]
Teacher (U23T) Oh, this is not very nice! Unfortunately I am very disappointed with your attitude. Michał [boy’s name, transl.], what
happened?!
Student (U24S11) this is impossible
Teacher (U25T) It is possible! Everything is possible!
Student (U26S11) No
Teacher (U27T) well, let me prove it to you that it is indeed possible
[Noise]
Student (U28S11) Than why can’t I fly?
Student (U29S12) or I?
(…)
Fig. 1 depicts the pattern of interactions between students and teachers for this segment of classroom discussion.

4.4.2. Micro-analysis of interactions in null classroom

In this classroom, the teacher presented the students with an open question and informed them that they would be conducting a
brainstorming session. The students were asked to come-up with as many unusual uses for a comb as they could (i.e., other than its
primary function). The first student shared an idea (U2S1) that was reinforced by two other students (U3S2; U4S3) and, ultimately,
accepted (A) by the teacher (U5T). As depicted in Fig. 1, this interaction starts by following an open (indeterminate) trajectory,
wherein two other students endorse the first student’s idea, and then moves into a more closed (determinate) trajectory as the teacher
accepts (A) it by re-voicing it and thereby closes the interaction.
In that same utterance (U5T), the teacher initiated another interaction by calling on one of the two students (Tadek) who agreed
with the first student. This student shared his idea (U6S3), which was endorsed by another student (U6S4), but the teacher did not
acknowledge it and, instead, called on another student (U8T). Consequently, the Tadek’s idea (U6S3) remained suspended in the
indeterminate space of the interaction (see Fig. 3).

257
A. Gajda et al. Thinking Skills and Creativity 24 (2017) 250–267

Fig. 1. Teacher-Student Ideational Pathways in class with no correlation between creativity and school achievement identified. Note: U# = represents each utterance
in temporal order. T = teacher utterances. S# = student utterances. A = accepted, C = contested; D = dismissed, S = suspended; Dotted double arrow line = -
returning to an idea.

The next student (Kamila) who was called on by the teacher, provided another idea (U9S5), which was immediately contested by a
peer’s question (U10S6). The teacher calls on the student (Karolina) (U11T) who contested her peer. Karolina responded to this
invitation by sharing a different idea (U12S6), instead of explaining her point of contestation that she raised (U10S6) about Kamila’s
idea (U9S5). Consequently, Kamila’s idea and Karolina’s contesting of that idea were both suspended and not explored by the teacher
or other students (see Fig. 1).
Karolina’s new idea (U12S6) was immediately questioned and, ultimately, dismissed by the teacher (U13T). Another student
(U14S7) shared an idea, which was initially accepted by the teacher (U15T), but then contested by two students (U16S8, U17S6). Instead
of exploring this interaction further, the teacher moved on by calling on yet another student (Kacper) and thereby suspended the
interaction that was underway (U18T).
Kacper shared his idea (U19S9), which was left suspended as the teacher (U20T) immediately called on yet another student
(Weronika). Weronika shared her idea (U21S10), which was immediately questioned by a peer (U22S11), and a quarrel between the
children ensued. The teacher attempted to re-direct the nature of the discussion (U23T) and asked one of the students (Michał) what
happened. The student (S11) who questioned Weronika reanimated his contestation of Weronika (denoted by a dotted line in Fig. 3)
and re-voiced his protest about using a comb as a shish-kebab skewer (U24S11). The teacher then suggested that anything is possible
(U25T) and the student contested the teacher’s argument (U26S11). The teacher attempted to defend the feasibility of the idea (U27T),

Fig. 2. Teacher-Student Ideational Pathways in class with a negative relationship between creativity and school achievement. Note: U# = represents each utterance in
temporal order. T = teacher utterances. S# = student utterances. A = accepted, D = dismissed.

258
A. Gajda et al. Thinking Skills and Creativity 24 (2017) 250–267

Fig. 3. Teacher-Student Ideational Pathways in class with a positive relationship between creativity and school achievement. Note: U# = represents each utterance in
temporal order. T = teacher utterances. S# = student utterances. A = accepted, D = dismissed, S = suspended. The dotted lines (—) represent the teacher re-
animating and building on a previous ideas.

but the student challenged the teacher’s claim that anything is possible (U28S11) and received support from another student (U29S12)
and ultimately drove the interaction in a closed direction (see Fig. 1).
In considering this pattern of interaction, we see the teacher inviting student ideation, but the discussion seemed to lack the
structure necessary for clarifying the expectations or goals for the brainstorming session. Without such structure, a discussion can
drift into a more chaotic and disruptive pattern of interaction (Beghetto & Plucker, 2015). Indeed, the teacher did not provide
students with clear expectations for how to share their ideas or guidelines for how they should respond to the ideas of others.
Moreover, the teacher often did not explore or expand on the ideas or questions that students shared.
Two key components of creative learning – exploring and providing feedback on ideas (Beghetto, 2016b) – were largely absent
from the interaction. Students did contest ideas (including their teacher’s at the end of the session), but there was not adequate
interactional support or time for further exploration of ideas, particularly discrepant ideas (such as using the comb as a shish-kebab
skewer). Moreover, this pattern of interaction illustrates that it is not only teachers who sometimes take interactions into a closed
trajectory, but that one or two persistent students can also contest and stifle the teacher’s invitations to explore a peer’s
unconventional idea (Beghetto, 2016a). In sum, although the teacher provided opportunities for students to express and explore
unconventional ideas, the guidance and expectations necessary for exploring, developing, and refining ideas seemed to be lacking
(Beghetto, 2016b; Beghetto & Plucker, 2015).
A potential insight that can be gleaned from this brief micro-level analysis of a classroom classified as having a null relationship
between creativity and academic achievement is that even though teachers may provide opportunities for open ideation, additional
guidance seems necessary to develop those ideas into creative learning contributions that, in turn, may be reflected in connections
between creativity and academic learning. Whether similar interactional patterns occur in other classrooms with null relationships is
an empirical question that warrants subsequent analysis. Still, this kind of micro-level analysis offers researchers with an example of
how they might use similar diagrammatic approaches in conjunction with more macro-level, quantitative classifications to concretely
illustrate the kinds of dynamic patterns of interactions that can be depicted, interpreted, and explored in subsequent theoretical and
empirical work.

4.4.3. Teacher–student interaction in a negative classroom

The following transcript of a teacher-student interaction is from a classroom classified as having a negative relationship between
creativity and academic achievement (r = −0.34, n = 28). The interaction occurs as part of a 2nd grade mathematics lesson.
Teacher (U1T) How else do we call 24:00? 00 or…?
Student (U2S1) Twelve?
Teacher (U3T) Twelve at night, 24, or…?
Student (U4S2) Midnight.
Teacher (U5T) Midnight, well done Bartek [boy’s name, transl.]. We write down hours like this, look at me, 5:00–just like it is
screened on an electronic clock or with two dots, like we did in first grade, or with a dot. But when we mean an afternoon
hour, then we will write seventeen, 17:00, right? We said today, just a while ago, that a day is 24 h and it is made up of a
light part, how do we say this, when it is …
Student (U6S3) … bright outside.
Teacher (U7T) Day and night. Take a look, when we speak about hours during the day. Then we say 5 in the morning, 9 in the
morning, 12 noon, 15 in the afternoon or 20 in the evening. Meanwhile, night hours, 22 at night, 24 or midnight, one a

259
A. Gajda et al. Thinking Skills and Creativity 24 (2017) 250–267

night or even three at night. The second exercise, compare what the clocks show. The ordinary clock shows what time,
Agata [girl’s name, transl.]?
Student (U8S4) One.
Teacher (U9T) One. Do we know if it is one or thirteen hour? Or can we write it down as 13:00 when we mean day and then one
when it is…? One?
Student (U10S5) …00
Student (U11S6) …in the morning.
Student (U12S7) …at night.
Teacher (U13T) At night, what are you talking about, one in the morning, really? I am still asleep then. An electronic clock screens
thirteen hundred hours during the day and at night it shows one. And we know right away whether it is day or night. Let’s
assume that we are in some strange room and we don’t know if it is day or night, but that clock screens it for us and we
already know. Underneath the frames, please write “day” or “night.” Please read the first of the series of hours, Antek [boy’s
name, transl.].
Fig. 2 depicts the pattern of interactions between students and teachers for this segment of classroom discussion.

4.4.4. Micro-analysis of interactions in negative classroom

In this classroom, the teacher initiated the interaction by reviewing information about how to read hours in a 12-h and 24-h clock
system. The teacher initiates the interaction by asking a question (U1T), a student responds (U2S1), the teacher accepts the response
and initiates a new question, “Twelve at night, 24, or…?” (U3T). This pattern of interaction is depicted in Fig. 2 as an inverted U,
which involves the teacher initiating an opening for students to respond, a student responds to the invitation, the teacher provides a
closed (determinate) evaluation of that response, and immediately initiates a new question.
A different student (Bartek) responded to the teacher’s new question by providing the expected answer, “Midnight” (U4S2). As
depicted in Fig. 2, the interaction follows the similar inverted U shape, with the teacher accepting Bartek’s response (U5T) and then
initiating a new opening for student response by tacitly invited students to respond by completing her sentence, “the day is 24 h and it
is made up of a light part, how do we say this, when it is…”.
A different student attempted to complete the teacher’s sentence by saying, ‘…bright outside’ (U6S3). Once again, the similar
inverted U-shape pattern of interaction ensues (see Fig. 2), with the teacher moving the interaction into closed (determinate)
trajectory by dismissing (D) the students’ response, providing the answer she was expecting herself, and then initiating a new
question directed at a different student (U7T).
In this instance, the teacher initiated a new opening by directly calling on Agata to respond. The inverted U pattern is repeated
once again (see Fig. 2), with the student (Agata) providing a response (U8S4) that her teacher quickly accepted, re-voiced, and then
initiates a new opening in the form of an incomplete sentence, “One. Do we know if it is one or thirteen hour? Or can we write it down
as 13:00 when we mean day and then one when it is…? One?” (U9T).
Although several students provided responses to this invitation (U10S5; U11S6; U12S7), the same basic inverted U pattern of
interaction was repeated (see Fig. 2), resulting from the teacher dismissing all but the last student’s response (U12S7), which she
accepted and re-voiced (U13T). In this segment the teacher also took a moment to further discount one student’s idea (U11S6), by
saying in a somewhat mocking way “…what are you talking about, one in the morning, really? I am still asleep then” (U13T). The
segment closed with the teacher providing additional commentary about electronic clocks, directing students to label examples, and
asking a student to read aloud the a series of hours.
Our visual diagram (Fig. 2) illustrates a repeating pattern of briefly open and then rapidly closing interactions directed by the
teacher. This pattern of talk represents a very common three-part pattern of discourse described elsewhere (Beghetto, 2016c; Cazden,
2001; Goodlad, 2004; Mehan, 1979), which involves: the teacher asking known-answer questions, student’s attempting to guess the
expected response, and the teacher immediately evaluating whether the students’ responses match what is expected. Indeed, in the
case of the interactions analyzed above, the teacher quickly accepted expected responses, re-voiced them, occasionally provided
additional explanation, and used them to quickly move on to the next question. Any unexpected responses were dismissed either by
being simply ignored or, in one case, seemingly mocked by the teacher. Moreover, students did not have an opportunity to elaborate
on their ideas or provide feedback on each other’s ideas. Although such patterns of talk are not always problematic and can
sometimes be justified (e.g., quickly reviewing information), they can become problematic for supporting creative learning when they
represent the modal form of interaction in a classroom. Indeed, the lack of extended exploration of ideas and the lack of instructional
feedback provided to students – coupled with a mocking evaluation of students’ ideas – can have a chilling effect on students’
willingness to share, test-out, and develop their ideas and thereby stifle creative learning (Beghetto, 2016b).
Given that our depiction and interpretation of this pattern of interaction is based on a very brief segment from one particular
discussion, from one particular classroom, on one particular day; it would be inappropriate to make strong claims about this pattern
representing a typical pattern of discourse for this classroom let alone generalizing to other classrooms. Indeed, other factors may be
at play. It is possible that closed patterns of interaction tend to be more likely in particular subject areas (mathematics) versus other
subject areas (e.g., language instruction) – regardless of the relationship between creativity and academic achievement. It may also be
the case that the teacher felt pressured to quickly move through the material. As such, further exploration of the frequency and
conditions under which such patterns emerged is needed prior to making claims about this pattern being characteristic of particular
types of classrooms. This includes exploring the perspectives, rationale, and experiences of the teachers and students involved in such
interactions.
Still, even with such caveats in mind, we would assert that the results of our initial, exploratory efforts provide creativity

260
A. Gajda et al. Thinking Skills and Creativity 24 (2017) 250–267

researchers with a concrete, visual depiction of a closed, inverted U-shape pattern that can be used as a basis for subsequent empirical
testing and validation. Such efforts can help clarify whether this reoccurring inverted U-shape pattern of interaction is more likely to
occur in classrooms characterized as having a negative relationship between creativity and academic achievement. Subsequent work
may also clarify whether and how other factors (e.g., subject area, lesson goals, time constraints, teacher’s confidence with the subject
matter) mediate and moderate the occurrence and frequency of this particular pattern of classroom talk. We therefore encourage
researchers to explore the frequency and consistency of this finding in subsequent work.

4.4.5. Teacher–student interaction in a positive classroom

The following transcript of a teacher-student interaction is from a classroom classified as having a positive relationship between
creativity and academic achievement (r = 0.60, n = 22). The interaction occurs as part of a 6th grade Polish language lesson.
Teacher (U1T) (…) We have to find out what the word ‘universalism’ means. Maybe you know the adjective ‘universal?’
Student (U2S1) One of a kind
Teacher (U3T) One of a., meaning what?
Student (U4S1) One in its kind?
Teacher (U5T) Well, I don’t think so (high noise) … but let’s take it one-by-one: Robert comes first
Student (U6S2) Well, you know, meaning everything
Teacher (U7T) Everything, you are on the right track, right?
Student (U8S2) Well, you know, that they aren’t …
Teacher (U9T) For example with what…? but one-by-one, please, for example with what other words can ‘universal’ go together with?
Student (U10S2) Person
Teacher (U11T) What associations?
Student (U12S3) Car
Teacher (U13T) A universal car, meaning what kind?
Student (U14S4) it has many uses
Teacher (U15T) it has many uses, right?
Student (U16S5) A universal computer
Teacher (U17T) Computers can be universal
Student (U18S6) A remote control
Teacher (U19T) A universal remote control; what is it for?
Student (U20S6) To put the volume up or down
Teacher (U21T) For your garage door, TV, right, so it is for…? Please. And when we have a t-shirt, then it has a universal size, right?
Meaning what kind of size?
Student (U22S7) (high noise) One that fits…
Teacher (U23T) One that fits everybody. So, see? We have an idea about what ‘universal’ means. Let’s now try a noun, right, it is
‘universalism’ that is the subject of this lesson; what may it mean?
Fig. 3 depicts the pattern of interaction between students and the teacher.

4.4.6. Micro-analysis of interactions in positive classroom

In this classroom, the teacher initiates the interaction by asking students whether they know the meaning of the adjective
“universal” (U1T). A student provided a response, “one of a kind” (U2S1), which the teacher explored further by inviting the student to
explain what “one of a kind” means (U3T). The student then slightly modified the initial response and changed it from a statement to
a question (U4S1). The teacher contested this idea, “Well, I don’t think so…” and ultimately dismisses it – moving on to another
student, “…but let’s take it one-by-one: Robert comes first” (U5T). As depicted in Fig. 3, this pattern of interaction represents an
extended, inverted U-shape, which includes additional exploration and engagement with the student’s idea prior to dismissing it and
initiating a new pattern of interaction with a different student (Robert).
The next student, Robert, shares his idea, “Well, you know, meaning everything” (U6S2), which the teacher re-voiced and invited
additional elaboration, “Everything, you are on the right track, right?” (U7T). As Robert attempted to clarify, “Well, you know, that
they aren’t…” (U8S2). The teacher interrupted Robert and took the discussion in a different direction by initiating a new question, “…
what other words can universal go together with?” (U9T). As depicted in Fig. 3, this exchange left Robert’s initial idea suspended.
Robert was the first to respond to the teacher’s new question, offering: “Person” (U10S2). The teacher did not acknowledge
Robert’s new idea – again leaving it suspended. Instead, the teacher posed a leading question, “What association?” (U11T).
A different student provided a response “Car” (U12S3) that the teacher used as the basis for a question, “A universal car, meaning
what kind?” (U13T). Yet another student offered a response, “it has many uses” (U14S4), which the teacher accepted by re-voicing, “it
has many uses” and then added the question “right?” to invite further exploration (U15T).
This extended exchange amongst the teacher and two students is depicted in Fig. 3, including how this exchange invited yet
another student (S5) to offer a new idea “A universal computer”(U16S5), which the teacher accepted by re-voicing it, “Computers can
be universal” (U17T). A moment later a different student provided an idea, “a remote control” (U18S6), which the teacher re-voiced,
“A universal remote control” and invited further exploration by asking “what is it for?” (U19T). The same student who offered
“remote control” responded by explaining that it can be used “to put the volume up or down” (U20S6). The teacher built on that
response and used it to invite more responses “For your garage door, TV, right, so it is for…? Please.” Before students could respond,
however, the teacher initiated two additional questions, “And when we have a t-shirt, then it has a universal size, right? Meaning

261
A. Gajda et al. Thinking Skills and Creativity 24 (2017) 250–267

what kind of size?” (U21T). A different student started responding by saying “One that fits…”(U22S7), which the teacher picked up and
finished, “One that fits everybody” (U23T). The teacher concluded the interaction by explaining that the group now has an idea of
what universal means, and then directs the students to explore the meaning of the noun, “universalism.”
This more extended and linked exchange amongst the teacher and several students is depicted in Fig. 3 and we use dotted lines to
illustrate how ideas shared at earlier points in the exchange were re-animated and built upon to develop a growing, more emergent
and collective understanding. Such an exchange seems to be consistent with what would be expected in a more improvisational or
participatory form of creative collaboration (see Clapp, 2017; Hanson, 2015; Sawyer, 2010). Indeed, as Sawyer (2010) has argued,
creativity in groups is characterized by “collaborative emergence.” This form of creative expression is emergent because the “outcome
cannot be predicted in advance” and it is collaborative because “no single participant can control what emerges – the outcome is
collectively determined by all participants” (Sawyer, 2010; p. 181).
Although the patterns of interactions in this excerpt from a positive classroom have some distinct features, there are also some
similarities with the patterns of interaction analyzed in the null classroom (e.g., inviting multiple student responses) and negative
classroom (i.e., re-voicing expected responses, ignoring responses and re-directing to expected responses). What seems to be different
in this classroom, however, is that the teacher explored and encouraged elaboration of student ideas and used those ideas in a more
open and emergent fashion (Beghetto, 2016b; Sawyer, 2010). Moreover, the teacher provided more guidance and structure and there
was less closure and no mocking of ideas as compared to the null and negative classroom.
Moreover, the teacher in this excerpt seemed to demonstrate a willingness to explore and try to understand the way students were
thinking. This is not to say that there were no instances of dismissing or ignoring student ideas. Indeed, there were instances when the
teacher cut-off students and attempted to direct students with leading questions and explanations. A potentially important difference,
however, was that teacher in this excerpt from a positive classroom often explored first, then evaluated or re-directed. This “explore
first, then evaluate” approach is one recommended way of how to promote creative learning (Beghetto, 2013, 2016b). It may be the
case, that classrooms with a positive association between creativity and learning tend to be characterized by a more exploratory
interactional pattern as illustrated in Fig. 3. Again, such claims require additional empirical exploration and verification.

5. General discussion

Prior to discussing our findings, we first want to highlight a few limitations of the study. Given that this article reports on an
exploratory and somewhat ambitious endeavor, we recognize that our approach has limitations that require additional refinement
and replication.
More specifically, our analysis was limited by the use of teacher grades to select classrooms, brief observations of single lessons in
the ten classrooms selected, and our inability to control for other potentially important moderating and mediating factors.
Consequently, subsequent efforts would benefit from using larger, more diverse samples and observations across time and across
grade levels and subject areas. Also, given that the expression of creative thought and action can be variable across different subject
areas (Baer, 2016), different ages of students (Gajda et al., 2017), different teachers (Beilock, Gunderson, Ramirez, & Levine, 2010),
and different measures of creativity and academic achievement (Gajda et al., 2017); it will be important that researchers examine
these potential mediating and moderating factors in subsequent efforts.
Moreover, although we used systematic observation protocols and the resulting behavioral factors demonstrated good internal
consistency, our use of a single observer may have also limited the reliability of these observations. We therefore recommend that
subsequent efforts include video recordings of lessons, which can be coded and cross-checked by multiple observers. Similarly, our
micro-level analyses and visual depiction of excerpts from lessons were based on our retrospective analyses. A more robust approach
would include sharing our analyses with teachers (and even students) in an effort to further explore and understand what additional
factors and constraints (e.g., temporal, socio-emotional, physical, situational) may have been at play and thereby influenced the
interactional responses and patterns depicted herein (Beghetto, 2016a). We therefore recommend that researchers consider ways of
assessing the perspectives and experiences of teachers and students using methods such as experience sampling (Hektner,
Schmidt, & Csikszentmihalyi, 2007).
Still, even with these limitations, we feel that this initial effort represents a promising and much needed multi-method approach
for exploring the dynamic nature of creative learning in the classroom. In what follows we will briefly discuss some key findings from
this initial effort and offer some concluding thoughts.

5.1. Teacher and student behaviors

Given that a classroom is a system (Parsons, 2008), teachers and students dynamically shape how that system operates and is
experienced by virtue of the behaviors they exhibit. With respect to creative learning, the classroom environment plays a key role in
determining whether creative learning will be supported or suppressed (Beghetto, 2016b). Our results indicate that there were
several types of teacher and student behaviors that seemed to vary as a function of whether classrooms had a positive association with
creativity and learning (versus a negative or null relationship).

5.1.1. Teacher behaviors

Teachers in classrooms with a positive association tended to demonstrate behaviors that we labeled as more “caring” (i.e., careful
listening, authenticity of the teacher’s behaviors, delayed assessment of ideas) and “emotionally supportive” (e.g., friendly humor,
motivating to manage failure). These findings align with prior work that has highlighted the importance of the affective support when

262
A. Gajda et al. Thinking Skills and Creativity 24 (2017) 250–267

it comes to behaviors associated with supporting creative learning (Kader, 2008; McGreevy, 1990). Indeed, this type of emotional
scaffolding (Rosiek & Beghetto, 2009) seems to not only support creativity and learning, but in the context of our findings, seems to
play a larger role than behaviors associated more directly with supporting creativity (e.g., encouraging creativity and providing risk
acceptance). Although these findings require further exploration in subsequent studies, they do suggest that establishing an
emotionally supportive and caring environment may provide the psychological safety necessary for students to take the risks required
for creative learning (Beghetto, 2016b).
Conversely, in classrooms with null and negative relationships between creativity and learning there seemed to be limited
opportunities to share and build on ideas. Moreover, there seemed to be greater instances of quick acceptance, dismissals, and
contesting of ideas. Although contesting ideas can be an important aspect of providing creativity supportive feedback, how that
feedback is delivered and received can play an important role in whether it supports or suppresses creative expression (Beghetto,
2007, 2016b; Niu, 2007).
Another interesting set of findings with respect to teacher behaviors pertained to lesson duration. Specifically, our findings
suggest that some behaviors such as encouraging creativity and emotional support may be more difficult for teachers to sustain across
the duration of a lesson. To the extent that these behaviors are important for encouraging students’ creative thought and behavior in
the classroom, it will be important to examine the consistency and consequences of these findings in subsequent efforts.

5.1.2. Student behaviors

With respect to student behaviors, our results indicate that there were not many differences across types of classrooms. One
pattern of findings that is interesting, however, is that positive engagement tended to decline across all classroom types and students
in classrooms classified with a negative relationship between creativity and achievement, tended to demonstrate lower levels of
misbehavior, self-expression, and positive engagement. These findings suggest that classrooms classified with a negative relationship
between creativity and achievement may reinforce conformity to teachers’ expectations – both cognitive (i.e., provide expected
responses) and behavioral (i.e., behave in expected ways). Whereas classrooms with a null relationship may occasionally result in
higher levels of misbehavior because, as we noted in our micro-analysis of interactions, there may be a lack of necessary structure and
guidance provided by the teacher.
These findings offer new and more nuanced insights into previous work that has tended to associate student creativity with
perceived misbehavior (Stone, 1980; Westby & Dawson, 1995). Specifically, creativity may be able to co-exist with academic learning
and behavioral expectations as long as there is sufficient structure and guidance in place. However, students may lack the stamina to
maintain high levels of positive engagement during the course of a 45-min lesson. Again, these findings point to directions for
subsequent study by creativity researchers.

5.1.3. Relationship between teacher and student behaviors

Regardless of classroom type, we found that encouraging creativity was associated with students’ demonstrating positive
engagement, self-expression, and ideation. These results contribute to and align with previous work, which has suggested that
actively encouraging creativity can result in behaviors associated with creative expression (Davies et al., 2012;
Feldhusen & Treffinger, 1980; Torrance, 1962). Teacher’s risk acceptance was also positively related to students’ self-expression
and ideation. As we mentioned above, however, encouraging creative expression and risk acceptance were not more pronounced in
classrooms that demonstrated a positive correlation between creativity and learning. Indeed, our findings suggest that demonstrating
care and emotional support may be more important to establishing creative learning environments that more direct forms of
encouraging creativity. These findings raise some interesting questions, particularly as they pertain to the idea that creative learning
benefits from the productive conflict of ideas and encounters with difference (Beghetto, 2016b; Glăveanu & Beghetto, 2016),
including: Are there certain ages, topics, or situations when the experience of care and emotional support are more important for
promoting the productive conflict of ideas? Do younger students, for instance, benefit more from such conditions than older students?
Or might there be a certain threshold of caring behaviors and emotional support necessary for creative learning to occur? These
questions warrant further exploration by creativity researchers.

5.2. Discussion of teacher and student interactions

The results of our more micro-focused analyses of teacher and student interactions highlight how using concrete, visual displays of
interactions – even brief interactions – can reveal potentially important insights into the norms and expectations of different types of
classrooms (Beghetto, 2016a; Tanggaard & Beghetto, 2015). As we discussed, classrooms supportive of creative learning seem to be
those characterized by patterns of interaction that provided openings for students to share their ideas and opportunities to further
explore those ideas (even when expected answers are provided). This differs from classrooms where openings are provided, but not
explored or receive more closed and directive evaluations (negative and null classrooms).
This is not to say that directive and even dismissive moments are completely absent from the classroom with a positive association
between creativity and academic learning. Indeed, our analysis highlighted moments of dismissal, suspended (abandoned) ideas, and
directives by the teacher. These findings suggest seemingly stifling behaviors may not always be problematic, but that the frequency
or intensity of such behaviors may be more of an important factor to consider in classrooms that are more or less supportive of
creative learning. Again, such speculations require further empirical exploration.
Moreover, as we have discussed, our findings indicate that regardless of classroom type (positive, negative, null) it may be
difficult for teachers and students to sustain behaviors associated with creative learning over the duration of the entire lesson. One

263
A. Gajda et al. Thinking Skills and Creativity 24 (2017) 250–267

reason may be that as the lesson progresses, teachers may feel the pressure to move students to predetermined learning goals and
outcomes. Moreover, students may start to lose interest as the lesson progresses for more than 15 min.
These findings highlight the dynamic and more nuanced nature of classroom interactions and suggest to us that no classroom is
monolithic with respect to how teacher or student behaviors are displayed at different times across different patterns of interactions.
Still, given that we did note differences in the behaviors and interactions across the three classroom types, there may be consistent
and salient features of those behaviors (e.g., demonstrating care and providing emotional support) and interactions (e.g., inviting
elaboration shared ideas) that established classroom conditions more conducive to creative learning.
Along these lines, there are several important questions to be explored in subsequent work, including: Is there an optimal balance
that can be struck between exploratory versus directive teaching? How might that balance vary by academic subject area and sub-
topics within subject areas? Are there more or less optimal durations of a classroom interactions when it comes to supporting creative
learning? How might teachers who are more closed and directive in their approach incorporate more opportunities for open
exploration? Conversely, how might teachers who take a more open approach find ways to provide necessary structure, feedback, and
directives to keep the lesson moving along? Under what conditions might productive conflict of ideas be best promoted – does this
vary by age, subject matter, and participants in the room? What other moderating and mediating factors might be play in establishing
conditions more or less supportive of both creativity and learning?

6. Conclusion

Creative learning is a complex process that is not easy to examine in a single study, even one that uses multiple methods.
However, we hope that the blended approach we have demonstrated in this study provides researchers with insights into how they
might use a combination of quantitative and qualitative methods to more thoroughly examine creativity and learning in classroom
settings. We therefore encourage researchers to test the consistency of our findings reported herein and build on these findings in
their subsequent work. Although studies such as this one are time and resource intensive, we believe they have the potential to
deepen our understanding of how something as complex as creative learning might be better understood and, ultimately, supported in
the classroom.

Appendix A. Student behavior examples

Item Example of interaction or behavior

Questioning classroom rules S: Miss, can I change my seat? because I'm not comfortable here
T: You know the rules; we do not change our seats during the lesson
S: (…) but really I'm not comfortable here, but can I?
T: No, you can't
S: But I want another seat!
T: Why are you turning away, do it yourself [task]
T: But it itches, it itches … and what are you laughing at??[to his classmates]
Behavioral risk T: Olek [name] Your behavior is inappropriate. Please improve
S: That’s not true! I will not!
T: But you know what is the risk? I'll have to give you a reprimand in your report card
Distraction of attention i.e. students are losing concentration during a brainstorming session. They start the sentence but
many distractors prevent them from finishing the speech
Inappropriate answers T: It is possible! Everything is possible!
S: no
T: well, let me prove it to you (…)
S: Then why can’t I fly?
Not being discouraged by setbacks T: Jaś [name], what did you learn during today's lesson? The last condition that you met? You
need to remember?
S: That the sum of two…er…, I don’t know .
T: Try again. What did you learn, what is the conclusion from this? That each interior angle of
a triangle, the triangle has three interior angles. What is the conclusion?
S: That the sum of these angles is always 180°
Transforming the student’s own and T: What is the problem we have in the book?
other students’ ideas S1: The Stranger (The main character of the book)
T: But please name the problem. Say what is the problem?
S2: No one is interested in him
Expressing emotion A student wants to tell an idea for a solution to the problem, after lesson is over:
T: I want to finish a lesson
S: and I want to tell my idea
T: And I want to finish a lesson!

264
A. Gajda et al. Thinking Skills and Creativity 24 (2017) 250–267

S: It's not fair!

Curiosity and interest in the topic Students carefully listen to the teacher who presents a new topic of a lesson:
T: And you shall know soon what you have to do in a group class. Soon I'll show you
S1: Cool!
T: here it is the hidden password [shows the board with mathematical tasks], what else can you
do in a group, in a class? What are we going to do? you will be guessing some things. I would
like to .
S2: I know, I know! Thirty! [gives the answer to the first task]
T: Who will guess the second task?
S: [students chorus] Me, me! I can do it!
Opposing the teacher’s views T: unfortunately, unfortunately (….) We don’t go to the court. After Maciek’s [the name] last
fall it might still better … Isn’t it?
S: It actually it isn’t better (…)
Engagement in new tasks i.e. students start to create mind maps (new task) with the willingness and curiosity
Ease in recalling words and T: So you say the characteristics of the comb well. Olive [name]?
expressions S1: This is handy
T: is a small handheld, it can be e.g. hidden in a purse or even in your pocket, Ewa [name]
S2: useful, helpful
T: Great! Julia [name]
S3: smooth, comfortable
Expressing one’s own views T: Please listen to the questions, and what our colleagues have to say, first read the question:
S: How do you think, why did the girls fail to complete the project? [gives an answer]. They
have to cope alone because Natalka [character in text] went to the store well and asked for a
shoe box and well, they need to deal with themselves.
Jokes and sense of humor [Discussion about how to encourage colleague to clean]
S1: it is better to clean up, because at the same time it can be fun to play. Throw it, play with it,
e.g. some missions up there, or collect blocks, disarm bombs and so on. It's fun.
S2: Unless you stumble finally, and such fun will come out of it
[laughs]
Many ideas from various categories i.e. Students give the use of term “universal” for the remote control, clothes, computers, etc.
Many answers to the same question i.e. Students give a wide variety of ideas on the question of the use of the comb.
Surprising answers T: What questions can we ask in connection with our text?
S: Did grandmother have hearts cut in the shutters?
[there is no such information in the text read in class]
Cordiality toward other students Pupils care for each other, in a situation where a student does not understand task, a colleague
offers him help

Appendix B. Teacher behavior examples

Item Example of interaction or behavior

Careful listening to questions During the discussion about the term “universality” a question not related to the subject of the lesson
is asked:
S: Miss, Can I borrow an eraser from you?
T: And why do you need it? We have not used a pencil so far
S: I know, but I am writing with a pencil now
Delayed assessment of ideas i.e. While giving answer to the question on the use of a comb by students, teacher does not evaluate
ideas for good and bad. She defers assessment and allows for discussion on the merits of each idea at
a later stage of lesson.
Group work organization T: Now, working in groups, solve problems as a group, as a team. That is: not one person solves the
problem, but all together
Authenticity, genuineness T: I'm sorry you could not appreciate the efforts of your colleagues (…)
Encouraging to search for diverse T: The task is to come up with as many uses for a comb but other than its basic application which is
ideas combing.
Encouraging to create many ideas T: Please note, those of you who generate the most of these ideas, this group [while working in
groups] will get rewarded. But remember, all the questions …. All the answers should address this
question.
Open questions T: How could the main character feel at that time? (During classes about one of the school books)
Encouraging divergent thinking T: Look for a variety of ideas in your heads, such that no one else would come up with

265
A. Gajda et al. Thinking Skills and Creativity 24 (2017) 250–267

Atmosphere of creative play i.e. During the classes, when possible, the teacher allows a playful atmosphere, playing with ideas
and moving away from rigid class rules.
Friendly humor T: There are three [about numerical tokens] Magda [name], you have three? that's good. Come on,
give one − all the same, after all they are not signed. Please. And what? Yours are missing? [to
another student] Is Dinosaur camouflaged or what? [about an illustration on the pencil case one of
the students]
Using constructive criticism T: What a great idea! It's not exactly the answer to our question. But how can this idea be developed,
improved?
Motivating to manage failures while solving a mathematical problem:
T: Sara [name]. What is the answer?
S1: The 3/10 x.
T: Victor?
S2. 3/10 x.
T. And the last group?
S2. 7/10 x. [Gives the wrong answer]
T: See, here the answer is wrong, then try to resolve it once again, can you see where was the
mistake?
Sensitivity to feelings T: Okay, folks. We read a book, you have heads now certainly full of emotion and I would like that
we begin to try to define… … er … what is the concern of the book. So, cruel notion that carries a lot
…. a lot of really moving to the depths of emotion.
Encouraging to take risk To a student who doubt whether to approach to mathematical competition:
T: What worries you? What can happen if you do?
S: Well, I do not know. I'm afraid…emm… that I’ll not be able to handle it
T: And how would you feel if you tried, and if you did well?
S: Cheerful
T: And if it went bad?
S: I would be nervous that I didn’t succeed.
T: It is your decision, but perhaps it… it is worth the risk?
Making it possible to choose and discussion on the use of extra time in class:
make decisions T: Would you prefer to play the math bingo now [game combines bingo and ability to perform
mathematical operations] or the intelligence? [coming up with answers that start with a particular
letter]
Acceptance of unconventional and T: I think the creativity in presentation [is noticeable], especially because not all the answers were
strange ideas consistent with the theme. But the boys here were tempted by a little bit of acting and it worked out
fine, right?
Assertiveness, respecting students’ T: We can’t change the date of the test. It is important to me that it’s held in a week, because then
feelings there is a long weekend and it will be harder to concentrate when you return. But I'm sure that you
will do well, because I see how hard you have worked so far.
Encouraging to test new ideas S: Miss, can I put it this way? [creating triangles of sticks of different lengths]
T: Well, check if that will suit
Tolerance of otherness and new T: hmm it’s an interesting point of view. I'm very curious of your opinion about Zenek Wojcik
ideas [literature character]
Encouraging to imagine different T: Close your eyes and try to see the girl, the character of a fairy tale: What is she wearing? How
things does she look? What is she doing? Where is she? What is she going through?

References

Andiliou, A., & Murphy, K. P. (2010). Examining variations among researchers' and teachers' conceptualizations of creativity: A review and synthesis of contemporary
research. Educational Research Review, 5, 201–219.
Baer, J. (2016). Content matters: Why nurturing creativity is so different in different domains. In R. A. Beghetto, & B. Sriraman (Eds.), Creative contradictions in
education: Cross disciplinary paradoxes and perspectives. Switzerland: Springer.
Barron, F. (1967). Creative person and creative process. New York: Holt, Rinehart and Winston.
Beghetto, R. A., & Plucker, J. A. (2015). Creativity and classroom management. In W. G. Scarlett (Ed.), Classroom management: An a-to-z guide. Thousand Oaks,
California: Sage.
Beghetto, R. A. (2007). Ideational code-switching: Walking the talk about supporting student creativity in the classroom. Roeper Review, 29, 265–270.
Beghetto, R. A. (2013). Killing ideas softly? The promise and perils of creativity in the classroom. Charlotte, NC: Information Age Press.
Beghetto, R. A. (2016a). Creative openings in the social interactions of teaching. Creativity: Theories-Research-Applications, 2, 261–273.
Beghetto, R. A. (2016b). Creative learning: A fresh look. Journal of Cognitive Education and Psychology, 15, 6–23.
Beghetto, R. A. (2016c). Leveraging micro-opportunities to address macroproblems: Toward an unshakeable sense of possibility thinking. In D. Ambrose, & R. J.
Sternberg (Eds.), Creative intelligence in the 21 st Century (pp. 159––174). The Netherlands: Sense Publishers.
Beilock, S., Gunderson, E., Ramirez, G., & Levine, S. (2010). Female teachers' math anxiety affects girls' math achievement. Proceedings of the National Academy of
Sciences, 107(5), 1860–1863. http://dx.doi.org/10.1073/pnas.0910967107.

266
A. Gajda et al. Thinking Skills and Creativity 24 (2017) 250–267

Cazden, C. B. (2001). Classroom discourse: The language of teaching and learning. Portsmouth, NH: Heinemann.
Chan, D. W., & Chan, L. (1999). Implicit theories of creativity: Teacher's perception of student characteristics in Hong Kong. Creativity Research Journal, 12, 185–195.
Clapp, E. P. (2017). Participatory creativity: Introducing access and equity to the creative classroom. New York: Routledge.
Cropley, A. J. (2001). Creativity in education and learning, a guide for teachers and educators. London and Sterling: Kogan Page.
Davies, D., Jindal-Snape, D., Collier, C., Digby, R., Hay, P., & Howe, A. (2012). Creative learning environments in education: A systematic literature review. Thinking
Skills and Creativity, 8, 80–91.
Dewey, J.,(1899/2007). The school and society. New York, NY, Cosmo.
Dewey, J.,(1934/2005). Art as experience. New York, NY, Perigee Books.
Esquivel, G. (1995). Teacher behaviors that foster creativity. Educational Psychology Review, 7, 185–202.
Feldhusen, J. F., & Treffinger, D. J. (1980). Creative thinking and problem solving in gifted education. Dubuque, IA: Kendall/Hunt.
Freund, P. A., & Holling, H. (2008). Creativity in the classroom: A multilevel analysis investigating the impact of creativity and reasoning ability on GPA. Creativity
Research Journal, 20, 309–318.
Günçer, B., & Oral, G. (1993). Relationship between creativity and nonconformity to school discipline as perceived by teachers of Turkish elementary school children,
by controlling for their grade and sex. Journal of Instructional Psychology, 20, 208–214.
Gajda, A., Karwowski, M., & Beghetto, R. (2017). Creativity and academic achievement: A meta-analysis. Journal of Educational Psychology, 109, 269–299.
Gajda, A. (2016). The relationship and moderators of school achievement and creativity at different educational stages. Thinking Skills and Creativity, 19, 246–259.
Glăveanu, V., & Beghetto, R. A. (2016). The difference that makes a creative difference. In R. A. Beghetto, & B. Sriraman (Eds.), Creative contradictions in education:
Cross-disciplinary paradoxes and perspectives. Switzerland: Springer.
Goodlad, J. L. (2004). A place called school: Prospects for the future. New York: NY: Mc-Grall-Hill.
Gralewski, J., & Karwowski, M. (2012). Creativity and school grades: A case from Poland. Thinking Skills and Creativity, 7, 198–208.
Gralewski, J., & Karwowski, M. (2013). Polite girls and creative boys? Students’ gender moderates accuracy of teachers’ ratings of creativity. Journal of Creative
Behavior, 47, 290–304.
Gralewski, J., & Karwowski, M. (2016). Are teachers' implicit theories of creativity related to the recognition of their students' creativity? Journal of Creative Behavior.
http://dx.doi.org/10.1002/jocb.140.
Guilford, J. P. (1950). Creativity. American Psychologist, 5, 444–454.
Guilford, J. P. (1967). Creativity: Yesterday, today: and tomorrow. Journal of Creative Behavior, 1, 3–14.
Hanson, M. H. (2015). Worldmaking: Psychology and the ideology of creativity. New York: Palgrave Macmillan.
Hektner, J., Schmidt, J., & Csikszentmihalyi, M. (2007). Experience sampling method. Thousand Oaks, California: Sage Publications.
Hong, E., Hartzell, S. A., & Greene, M. T. (2009). Fostering creativity in the classroom: Effects of teachers’ epistemological beliefs, motivation, and goal orientation.
Journal of Creative Behavior, 43, 192–208.
Hu, L. T., & Bentler, P. M. (1999). Cutoff criteria for fit indexes in covariance structure analysis: Conventional criteria versus new alternatives. Structural Equation
Modeling, 6, 1–55.
Jellen, H., & Urban, K. K. (1986). The TCT-DP (Test for Creative Thinking – Drawing Production): An instrument that can be applied to most age and ability groups.
Creative Child and Adult Quarterly, 11, 138–155.
Johnson, L. G., & Hatch, J. A. (1990). A descriptive study of four highly original preschoolers' creative and social development. Journal of Creative Behavior, 24,
205–224.
Kader, A. (2008). Enhancing creative behavioral expressions in school systems: The need for educational reform and center for creativity. The Pennsylvania State University,
Proquest, Umi Dissertation Publishing.
Karwowski, M., & Gralewski, J. (2013). Threshold hypothesis: Fact or artifact? Thinking Skills and Creativity, 8, 25–33.
Karwowski, M., & Jankowska, D. M. (2016). Four faces of creativity at school. In R. A. Beghetto, & J. C. Kaufman (Eds.), Nurturing creativity in the classroom (pp. 337–
354). New York: Cambridge University Press.
Karwowski, M. (2007). Teachers' nominations of students' creativity: Should we believe them? Are the nominations valid? Social Sciences, 2, 264–269.
Karwowski, M. (2017). Subordinated and rebellious creativity at school. In R. A. Beghetto, & B. Sriraman (Eds.), Creative contradictions in education (pp. 89–115).
Switzerland: Springer.
Kawenski, M. (1991). Encouraging creativity in design. Journal of Creative Behavior, 25, 263–266.
Kline, R. B. (2011). Convergence of structural equation modeling and multilevel modeling. In M. Williams, & W. P. Vogt (Eds.), Handbook of methodological innovation
in social research methods (pp. 562–589). London: Sage.
MacClelland, D. (1975). Strategies for effective teaching: A basis for creativity. AORN Journal, 22, 282–284.
McGreevy, A. (1990). Tracking the creative teacher. Momentum, 21.
Mehan, H. (1979). Learning lessons: Social organization in the classroom. Cambridge, MA: Harvard: University Press.
Muthén, L.K., Muthén, B.O.,(1998–2014). Mplus user's guide, 7th ed., Los Angeles, CA, Muthén & Muthén.
Niu, W. (2007). Western influences on Chinese educational testing. Comparative Education, 43(1), 71–91.
Nussbeck, F. W., Eid, M., & Lischetzke, T. (2006). Analysing multitrait–multimethod data with structural equation models for ordinal variables applying the WLSMV
estimator: What sample size is needed for valid results? British Journal of Mathematical and Statistical Psychology, 59, 195–213.
Parsons, T. (2008). The school class as a social system. In J. H. Ballantine, & J. Z. Spade (Eds.), Schools and society: A sociological approach to education. Los Angeles: Pine
Forge Press.
Ripple, R. E. (1977). Communication, education, and creativity. Contemporary Educational Psychology, 2, 219–231.
Rosiek, J., & Beghetto, R. A. (2009). Emotional scaffolding: The emotional and imaginative dimensions of teaching and learning. In P. A. Schutz, & M. Zembylas (Eds.),
Advances in teacher emotion research: The impact on teachers' lives. New York: Springer.
Sawyer, R. K. (2010). Learning for creativity. In R. A. Beghetto, & J. C. Kaufman (Eds.), Nurturing creativity in the classroom (pp. 172–190). New York: Cambridge
University Press.
Schacter, J., Thum, Y., & Zifkin, D. (2006). How much does creative teaching enhance elementary school students' achievement? Journal of Creative Behavior, 40,
47–72.
Scott, C. L. (1999). Teacher's biases toward creative children. Creativity Research Journal, 12, 321–328.
Soh, K. (2000). Indexing creativity fostering teacher behavior: A preliminary validation study. Journal of Creative Behavior, 34, 118–134.
Starko, A. J. (1995). Creativity in the classroom: Schools of curious delight. New York, NY: Longman.
Stone, B. G. (1980). Relationship between creativity and classroom behavior. Psychology in the Schools, 17, 106–108.
Tanggaard, L., & Beghetto, R. A. (2015). Ideational pathways: Toward a new approach for studying the life of ideas. Creativity. Theories-Research-Applications, 2,
129–144.
Torrance, E. P. (1962). Guiding creative talent. New York: Prentice-Hall Inc.
Urban, K., & Jellen, H. (1996). Test for creative thinking – drawing production (TCT-DP). Frankfurt: Swets Test Services.
Westby, E. L., & Dawson, V. L. (1995). Creativity: Asset or burden in the classroom? Creativity Research Journal, 8, 1–10.

267