Can a self-regulated flipped classroom improve creative performance? Evidence from a randomized controlled trial

ERCT Criteria Breakdown

• Level 1 Criteria

  • C

    Class-level RCT

    • Randomisation was carried out on individual students rather than on intact classes or schools, and the tutoring exception does not apply.
    • Fifty participants were randomly assigned to experimental or control groups using an online randomization tool (https://www.randomizer.org/) with a 1:1 allocation ratio.
    • Relevant Quotes: 1) "A randomized controlled trial was used to examine the effects of the flipped classroom on creativity. Fifty participants were randomly assigned to experimental or control groups using an online randomization tool (https://www.randomizer.org/) with a 1:1 allocation ratio." (p. 2) 2) "An initial pool of 283 students enrolled at the School of Art and Design was considered... After applying these criteria, 50 participants were retained." (p. 2) 3) "Intervention group (n = 25) - Received allocated intervention (n = 25) ... Control group (n = 25) - Received allocated intervention (n = 25)" (Figure 1, CONSORT flow diagram, p. 3) 4) "Additionally, as a single instructor was involved, teaching style and interaction patterns may have influenced the results. Even with controlled conditions, some degree of informal student interaction between groups cannot be entirely excluded." (p. 7) Detailed Analysis: Criterion C requires randomisation at the class level or stronger (school level), with an exception only for interventions that are inherently personal, such as one-to-one tutoring. Here the unit of randomisation was explicitly the individual student: fifty individually recruited volunteers were allocated 1:1 to the flipped or traditional condition using an online randomiser. No intact classes or schools were assigned. The intervention is a classroom pedagogy (flipped classroom with pre-class materials, in-class brainstorming, collaborative design projects), not personal tutoring, so the tutoring exception does not apply. The authors themselves acknowledge the contamination risk that class-level randomisation is meant to prevent, noting that "some degree of informal student interaction between groups cannot be entirely excluded" and that a single instructor taught both conditions. Criterion C is not met because randomisation was performed at the individual student level within a single institution rather than at the class or school level, and no tutoring exception applies.
  • E

    Exam-based Assessment

    • Outcomes were measured with psychometric creativity tasks (GAU, RAT) and a custom survey rather than with a recognised standardised exam.
    • Creativity was assessed before and after the intervention using the Guilford Alternate Uses Test and the Remote Associates Test.
    • Relevant Quotes: 1) "Creativity was assessed before and after the intervention using the Guilford Alternate Uses Test and the Remote Associates Test." (p. 1) 2) "Convergent thinking was measured using the remote associates test (RAT), first devised by Mednick (1962)." (p. 2) 3) "The Guilford alternate uses test (GAU) (Guilford, 1967) was used to evaluate divergent thinking." (p. 3) 4) "Participants were asked to consider two common objects, such as a toothbrush or a brick, and generate as many new and useful ways to use them as possible within 8 min (4 min for each object)... To avoid learning effects, the objects were changed to be a newspaper, a plastic bottle and an electric wire in the post-test." (p. 3) 5) "Responses were assessed by two trained independent raters who were blinded to group assignment. The raters were calibrated using a pilot dataset to ensure scoring consistency." (p. 3) 6) "Although the GAU and RAT are widely used and validated measures of divergent and convergent thinking, they primarily assess specific cognitive components rather than creativity in its entirety." (p. 7) 7) "The survey was aimed at measuring students' beliefs around their experience in the flipped classroom, with regard to SRL... The survey consisted of Likert scale items..." (p. 3) Detailed Analysis: Criterion E requires standardised, widely recognised exam-based assessment of educational outcomes - the kind of state-wide, national or curriculum-linked achievement exam that yields objective and comparable measures of school learning. The outcome instruments used here are the Guilford Alternate Uses Test and the Remote Associates Test. These are established psychometric laboratory tasks with published reliability evidence, but they are psychological measures of divergent and convergent thinking, not exam-based academic assessments of subject attainment. No course exam, national exam or standardised achievement test was administered. Moreover, the administration was locally adapted by the research team: the specific objects were chosen and changed between pre- and post-test, and scoring rules for fluency, flexibility, originality and elaboration were defined and applied by the study's own trained raters rather than by an external standardised scoring protocol. The third outcome, the self-reflection survey, is a custom Likert instrument. The authors also concede these measures "may not fully reflect real-world creative problem solving." Criterion E is not met because the outcomes were measured with psychometric creativity tasks and a custom survey, locally administered and scored, rather than with a recognised standardised exam.
  • T

    Term Duration

    • The post-test was administered 16 weeks after the intervention began, which equals roughly one full academic semester.
    • The flipped classroom intervention was implemented over a 16-week period.
    • Relevant Quotes: 1) "This study employed a 16-week self-regulated flipped classroom intervention to examine creativity outcomes, focusing on divergent and convergent thinking." (p. 1) 2) "The flipped classroom intervention was implemented over a 16-week period. Students were required to complete preparatory materials before class, while in-class time was devoted to interactive and creative activities designed to promote engagement and higher-order thinking." (p. 3) 3) "Creative performance (divergent and convergent thinking) was assessed at both pre-test and post-test, while the SRL survey was administered at the post-test to capture students' perceptions of their learning experience following the intervention." (p. 3) 4) "Further screening criteria included: (1) participation in at least one art or design course during the semester..." (p. 2) 5) "As the course progressed, students engaged in more complex tasks requiring them to apply creative thinking and technical skills, culminating in final project presentations and reflective discussions." (p. 4) Detailed Analysis: Criterion T requires that the primary outcome be measured at least one full academic term (approximately 3-4 months) after the intervention begins. The paper states repeatedly that the intervention ran for 16 weeks and that the post-test was administered after the intervention. Sixteen weeks is approximately four months and corresponds to a full semester; the paper frames the study as running across a "semester" course. The interval from intervention start to primary outcome measurement is therefore at least 16 weeks, which satisfies the one-term minimum. The measurement occurred immediately at the end of the intervention rather than at a delayed follow-up, but the standard permits this because the interval from intervention start to measurement is what matters. Criterion T is met because the intervention began and the post-test was taken 16 weeks (roughly one full semester) later, meeting the one-term minimum.
  • D

    Documented Control Group

    • The control group's size, attrition, baseline scores on all outcomes and business-as-usual condition are documented, though demographic detail is thin.
    • The experimental group received the flipped instruction, while the control group was taught using traditional methods.
    • Relevant Quotes: 1) "The experimental group received the flipped instruction, while the control group was taught using traditional methods." (p. 2) 2) "A randomized controlled trial was conducted with art and design students assigned to either an intervention group (n = 24) or a control group (n = 24)." (p. 1) 3) "Control group (n = 25) - Received allocated intervention (n = 25) - Did not receive allocated intervention (n = 0) ... Lost to follow-up (quit training) (n = 0) Discontinued intervention (n = 1) ... Analysed (n = 24)" (Figure 1, CONSORT flow diagram, p. 3) 4) "Fluency Pretest 3.864 +/- 0.089 [Intervention] 3.864 +/- 0.147 [Control] ... Originality Pretest 4.893 +/- 0.119 / 4.881 +/- 0.090 ... Flexibility Pretest 3.130 +/- 0.635 / 3.241 +/- 0.689 ... Elaboration Pretest 4.626 +/- 0.224 / 4.629 +/- 0.234 ... Convergent thinking Pretest 3.645 +/- 0.087 / 3.633 +/- 0.092" (Table 1, p. 5) 5) "An initial pool of 283 students enrolled at the School of Art and Design was considered. To ensure sufficient language proficiency, students who had not passed the College English Test Band 4 (CET-4) were excluded... Further screening criteria included: (1) participation in at least one art or design course during the semester, (2) no prior experience with flipped classroom instruction, and (3) willingness to participate in all course activities and assessments." (p. 2) 6) "In traditional lecture-based classrooms, learning can be sufficiently passive to provide fewer opportunities for creativity." (p. 4) Detailed Analysis: Criterion D requires the control group to be documented in terms of size, composition, baseline performance and the conditions it experienced. The paper reports the control group size at every stage of the CONSORT flow (25 allocated, 1 discontinued, 24 analysed), and Table 1 gives the control group's baseline (pre-test) means and standard deviations on all five outcome measures, so baseline comparability with the intervention arm can be directly assessed and is evidently close. The eligibility and screening criteria that define the population from which both arms were drawn are also stated (CET-4 pass, enrolled in an art or design course, no prior flipped classroom experience). The condition the control group received is stated: they "were taught using traditional methods", i.e. the lecture-based business-as-usual instruction described in the discussion, and no special treatment beyond normal teaching is indicated. The documentation is not exhaustive - no demographic breakdown (age, gender distribution, year of study) is given for either arm, and the traditional condition is not described session by session - but the size, baseline performance and treatment received by the control arm are all clearly quotable. Criterion D is met because the control group's size, attrition, baseline scores on every outcome and business-as-usual instructional condition are explicitly documented, despite the absence of a demographic table.

• Level 2 Criteria

  • S

    School-level RCT

    • The trial ran in a single institution with individual students as the unit of randomisation, so no school-level assignment took place.
    • This study was conducted at a single institution, which may limit the generalizability of the findings to other educational settings, institutions, or student populations.
    • Relevant Quotes: 1) "Fifty participants were randomly assigned to experimental or control groups using an online randomization tool (https://www.randomizer.org/) with a 1:1 allocation ratio." (p. 2) 2) "An initial pool of 283 students enrolled at the School of Art and Design was considered." (p. 2) 3) "This study was conducted at a single institution, which may limit the generalizability of the findings to other educational settings, institutions, or student populations." (p. 7) 4) "Additionally, as a single instructor was involved, teaching style and interaction patterns may have influenced the results." (p. 7) Detailed Analysis: Criterion S requires randomisation among schools, that is, among the institutions or units delivering the intervention. This study was carried out at one institution, with one instructor, and the randomisation was applied to individual student volunteers drawn from a single School of Art and Design. Only one implementing unit was involved, so school-level assignment was structurally impossible, and the authors explicitly acknowledge the single-institution limitation. Criterion S is not met because randomisation occurred at the individual student level within one institution, with no schools or sites randomised.
  • I

    Independent Conduct

    • The same author team designed, delivered, measured and analysed the intervention, with no external evaluator or third-party oversight.
    • Throughout the intervention, the instructor served mainly as a facilitator, guiding discussions and providing feedback rather than delivering traditional lecture-based instruction.
    • Relevant Quotes: 1) "YC: Conceptualization, Methodology, Writing - original draft. JF: Conceptualization, Data curation, Formal analysis, Investigation, Methodology, Writing - review & editing. YN: Formal analysis, Resources, Validation, Writing - review & editing. FT: Investigation, Methodology, Writing - original draft. LW: Data curation, Investigation, Software, Writing - review & editing. AM: Methodology, Resources, Supervision, Writing - review & editing." (p. 7) 2) "Pre-class materials included instructional videos, readings, and guided exercises, while in-class activities included brainstorming tasks (e.g., generating unusual uses for everyday objects), analysis of surrealist artworks (e.g., Salvador Dali and Rene Magritte), and collaborative design projects." (p. 4) 3) "Throughout the intervention, the instructor served mainly as a facilitator, guiding discussions and providing feedback rather than delivering traditional lecture-based instruction." (p. 4) 4) "Additionally, as a single instructor was involved, teaching style and interaction patterns may have influenced the results." (p. 7) 5) "Responses were assessed by two trained independent raters who were blinded to group assignment. The raters were calibrated using a pilot dataset to ensure scoring consistency." (p. 3) 6) "The author(s) declared that this work was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest." (p. 8) 7) "The author(s) declared that financial support was not received for this work and/or its publication." (p. 8) Detailed Analysis: Criterion I requires that the evaluation be conducted independently of those who designed the intervention, normally evidenced by an external evaluator, a third-party data collection agency, or an explicit statement separating the design team from the measurement and analysis team. Here the same author team conceptualised the study, designed and delivered the flipped classroom curriculum, administered the pre- and post-tests, curated the data and performed the formal analysis, as the CRediT statement shows. There is no external evaluation body, no independent data collection organisation, and no statement of third-party oversight; the ethics committee approval is a governance step, not an independent evaluation. The one element of independence is that the two raters scoring the divergent thinking responses were "trained independent raters who were blinded to group assignment." Blinded scoring reduces measurement bias in one outcome, but the raters were engaged and calibrated by the research team, and the design, delivery, data curation and statistical analysis all remained in-house. Blinded scoring alone falls short of the independent conduct that this criterion requires. Criterion I is not met because the authors designed, delivered, measured and analysed their own intervention with no external or third-party evaluation, blinded raters notwithstanding.
  • Y

    Year Duration

    • Outcomes were measured 16 weeks after the start, roughly one semester and well under 75 percent of an academic year.
    • This study employed a 16-week self-regulated flipped classroom intervention to examine creativity outcomes, focusing on divergent and convergent thinking.
    • Relevant Quotes: 1) "This study employed a 16-week self-regulated flipped classroom intervention to examine creativity outcomes." (p. 1) 2) "The flipped classroom intervention was implemented over a 16-week period." (p. 3) 3) "Creative performance (divergent and convergent thinking) was assessed at both pre-test and post-test, while the SRL survey was administered at the post-test to capture students' perceptions of their learning experience following the intervention." (p. 3) 4) "Future studies could also investigate the long-term sustainability of creativity gains through flipped classroom practice." (p. 7) Detailed Analysis: Criterion Y requires that outcomes be measured at least 75 percent of a full academic year (roughly 7 months or more of a 9-10 month year) after the intervention begins. The intervention here ran for 16 weeks, approximately 3.7 months or one semester, and the post-test was administered at the end of that period. There was no delayed follow-up: the authors explicitly leave the "long-term sustainability of creativity gains" to future research. Sixteen weeks is roughly 40 percent of a standard academic year, well short of the 75 percent threshold, and the shortfall is far greater than the minor (under 25 percent) deviation the standard tolerates. Criterion Y is not met because the total interval from intervention start to outcome measurement was about 16 weeks, far below 75 percent of an academic year.
  • B

    Balanced Control Group

    • The extra pre-class materials are integral to the flipped model being tested and reallocate rather than add class time, with class hours and instructor matched across arms.
    • By transitioning to pre-class content acquisition, students were afforded larger exposure during class to creative tasks, the consideration of alternative perspectives, and incremental refinement of thought processes that lie at the heart of divergent thinking.
    • Relevant Quotes: 1) "The experimental group received the flipped instruction, while the control group was taught using traditional methods." (p. 2) 2) "The flipped classroom intervention was implemented over a 16-week period. Students were required to complete preparatory materials before class, while in-class time was devoted to interactive and creative activities designed to promote engagement and higher-order thinking." (p. 3) 3) "It transfers information acquisition to pre-class activities and utilizes in-class time for active learning and promotes collaboration in the classroom, encouraging students to think critically and thereby improving capacities needed for divergent and convergent thinking." (p. 2) 4) "Instead of attributing the increased creativity observed in the end result to SRL as a direct mechanism, a flipped classroom also promotes creativity by redirecting instructional time into higher-order cognitive tasks (e.g., ideation, exploration, and in-class application). By transitioning to pre-class content acquisition, students were afforded larger exposure during class to creative tasks..." (p. 4) 5) "Pre-class materials included instructional videos, readings, and guided exercises, while in-class activities included brainstorming tasks... Students completed practical exercises using digital tools (e.g., Photoshop), with peer discussion and feedback to aid idea development and refinement." (p. 4) 6) "Additionally, as a single instructor was involved, teaching style and interaction patterns may have influenced the results." (p. 7) 7) "The author(s) declared that financial support was not received for this work and/or its publication." (p. 8) Detailed Analysis: Criterion B asks whether the intervention arm received additional time or budget that the control arm did not, and if so whether that addition is integral to the treatment being tested or a separable confound. Both arms were taught the same art and design course over the same 16-week semester by the same single instructor, so scheduled class contact time and staffing cost are matched. The intervention arm additionally received pre-class instructional videos, readings and guided exercises, plus in-class digital tool work; the control arm received conventional lectures with no reported parallel materials. This is an increase in materials and in out-of-class study demand for the intervention arm. Applying the decision tree: extra resources are present, but they are precisely the treatment variable. The flipped classroom is defined in this paper as the transfer of information acquisition to pre-class activity so that class time can be reallocated to active work - the authors describe it as "redirecting instructional time" and "transitioning to pre-class content acquisition" rather than adding instructional hours. The pre-class materials are not a separable add-on that could have been given to controls without dissolving the intervention; they constitute the flipped model itself. The control arm is the standard business-as-usual lecture comparison this design requires, and the study was unfunded, so no budget asymmetry is reported. Criterion B is met because the additional pre-class materials and digital exercises are integral to the flipped classroom model under test and represent a reallocation rather than an addition of instructional time, with the control arm serving as the appropriate business-as-usual comparison.

• Level 3 Criteria

  • R

    Reproduced

    • This 2026 study presents itself as filling a research gap and internet searching found no independent replication of its specific design.
    • Although the overall impact of flipped classrooms is well documented, little research explicitly examines the impact on creativity in art and design education.
    • Relevant Quotes: 1) "Very few studies have employed rigorous experimental methods such as randomized controlled trials to determine the causal effects of flipped learning on creativity." (p. 2) 2) "Although the overall impact of flipped classrooms is well documented, little research explicitly examines the impact on creativity in art and design education." (p. 2) 3) "This study was conducted at a single institution, which may limit the generalizability of the findings to other educational settings, institutions, or student populations." (p. 7) 4) "Effects of the flipped classroom instructional strategy on students' learning outcomes: a meta-analysis" (Cheng, Ritzhaupt and Antonenko, 2018, reference list, p. 8) 5) "From passive to active: the impact of the flipped classroom through social learning platforms on higher education students' creative thinking" (Al-Zahrani, 2015, reference list, p. 8) Detailed Analysis: Criterion R requires that this specific study be independently replicated by a different research team in a different context and published in a peer-reviewed journal. Internet verification was carried out in July 2026 against the Frontiers article record, general web search for the DOI 10.3389/fpsyg.2026.1807932, and searches for the author names combined with flipped classroom, creativity and replication terms. No replication study of this trial was found. Searches surfaced only thematically adjacent work by other teams, such as the quasi- experimental "The Impact of a Flipped Classroom on the Creativity of Students in a Cake Decorating Art Club" (Tien, Lin, Yin and Chang, 2020, Frontiers in Psychology) and Al-Zahrani (2015) on flipped learning and creative thinking. Neither is a replication of this 16-week self-regulated flipped design with GAU and RAT outcomes in art and design undergraduates; they use different designs, populations and measures. No verbatim quote claiming replication of this study could be located in any source, and none is invented here. The paper itself positions its contribution as filling a gap rather than confirming a replicated finding, and it was published on 29 May 2026 (corrected 12 June 2026), so an independent replication would in any case be implausibly fast. Criterion R is not met because no independent replication of this specific trial by a different research team could be found in any searched source.
  • A

    All-subject Exams

    • Criterion E is not met and the study measured only creativity constructs, with no attainment assessed in any curriculum subject.
    • This study employed a 16-week self-regulated flipped classroom intervention to examine creativity outcomes, focusing on divergent and convergent thinking.
    • Relevant Quotes: 1) "Creativity was assessed before and after the intervention using the Guilford Alternate Uses Test and the Remote Associates Test." (p. 1) 2) "This study employed a 16-week self-regulated flipped classroom intervention to examine creativity outcomes, focusing on divergent and convergent thinking." (p. 1) 3) "The survey was aimed at measuring students' beliefs around their experience in the flipped classroom, with regard to SRL... and creativity development." (p. 3) 4) "Although the GAU and RAT are widely used and validated measures of divergent and convergent thinking, they primarily assess specific cognitive components rather than creativity in its entirety." (p. 7) Detailed Analysis: Criterion A requires that impact be measured across all main subjects of the curriculum using standardised exam-based assessment, and it explicitly depends on criterion E: if E is not met, A cannot be met. Criterion E was judged not met because the outcomes are psychometric creativity tasks and a custom survey rather than standardised exams, so A fails on that dependency alone. Substantively, the study measured only creativity constructs - fluency, originality, flexibility, elaboration and convergent thinking - and no academic subject at all. Even the students' own art and design coursework attainment was not assessed, let alone other curriculum subjects, so possible spillover effects on other areas of study are entirely unmeasured. Criterion A is not met because criterion E failed and because the study assessed only creativity constructs rather than attainment across the main curriculum subjects.
  • G

    Graduation Tracking

    • Measurement stopped at the immediate post-test with no tracking to graduation in this or any locatable follow-up paper, and criterion Y was not met.
    • Future studies could also investigate the long-term sustainability of creativity gains through flipped classroom practice.
    • Relevant Quotes: 1) "Creative performance (divergent and convergent thinking) was assessed at both pre-test and post-test, while the SRL survey was administered at the post-test to capture students' perceptions of their learning experience following the intervention." (p. 3) 2) "The flipped classroom intervention was implemented over a 16-week period." (p. 3) 3) "Future studies could also investigate the long-term sustainability of creativity gains through flipped classroom practice." (p. 7) Detailed Analysis: Criterion G requires that participants be followed through to graduation from their educational stage, and it is conditional on criterion Y, which was not met here. Independently of that dependency, the paper reports no follow-up whatsoever beyond the immediate post-test at the end of the 16-week semester. The participants were university art and design students who were not tracked towards degree completion, and the authors explicitly flag long-term sustainability of the gains as a matter for future research. Internet verification was carried out in July 2026, searching for subsequent publications by Chen, Feng, Ning, Tang, Wang and Mohammadi that might track this cohort to graduation. No such follow-up publication was found in any searched source, and no verbatim quote evidencing graduation tracking exists to report; none is invented here. The May 2026 publication date makes any graduation follow-up implausible at this point. Criterion G is not met because criterion Y failed and because measurement stopped at the immediate post-test with no tracking towards graduation in this paper or any locatable follow-up publication.
  • P

    Pre-Registered

    • The paper reports institutional ethics approval only, with no trial registry identifier, pre-registration link or registration date.
    • This study was approved by City University Malaysia Research Ethics Committee (CITYU/RMC/R.ETHICS 2822) on 19 March 2025 and conducted in accordance with the Declaration of Helsinki.
    • Relevant Quotes: 1) "This study was approved by City University Malaysia Research Ethics Committee (CITYU/RMC/R.ETHICS 2822) on 19 March 2025 and conducted in accordance with the Declaration of Helsinki." (p. 2) 2) "A post hoc power analysis was performed using G*Power 3.1 (Heinrich Heine University Dusseldorf, Germany) for a repeated-measures ANOVA with a within-between interaction. The achieved power was 0.924, with alpha set at 0.05, a medium effect size (f = 0.25), and a total sample size of 48." (p. 2) 3) "The datasets presented in this study can be found in online repositories. The names of the repository/repositories and accession number(s) can be found in the article/Supplementary material." (p. 7) 4) "The studies involving humans were approved by City University of Malaysia Research Ethics Committee." (p. 7) Detailed Analysis: Criterion P requires that the full study protocol, including hypotheses, methods and planned analyses, be publicly pre-registered on a registry before data collection begins, with a verifiable identifier and date. The paper contains no reference to any trial registry - no ClinicalTrials.gov, ISRCTN, OSF, AEA or equivalent identifier, and no pre-registration date - anywhere in the Methods, the data availability statement or the ethics statement. What is provided is institutional ethics approval (CITYU/RMC/R.ETHICS 2822, 19 March 2025), which is a governance clearance rather than a public pre-registration of hypotheses and analysis plan. Internet verification in July 2026 confirmed against the Frontiers article record that no trial registration or pre-registration identifier is mentioned anywhere in the published article, so there was no registry entry to look up and no registration date to check against the start of data collection. The use of a post hoc rather than an a priori power analysis further indicates that the sample size and analysis plan were not fixed in advance in a registered protocol. Criterion P is not met because no pre-registration reference, registry identifier or registration date is given anywhere in the paper or in its online record.