Reimagining Science Education through Project-Based Learning: A Systematic Review of Creative Thinking Outcomes

Reimagining Science Education through Project-Based Learning: A Systematic Review of Creative Thinking Outcomes | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article Reimagining Science Education through Project-Based Learning: A Systematic Review of Creative Thinking Outcomes Imam Samodra, Fitria Rahmawati, Baskoro Adi Prayitno This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-8231968/v1 This work is licensed under a CC BY 4.0 License Status: Posted Version 1 posted You are reading this latest preprint version Abstract Fostering students’ creative thinking is a central goal of contemporary science education, and Project-Based Learning (PBL) has been widely promoted as a promising pedagogical approach to achieve it. The overall magnitude and consistency of PBL’s impact on creative thinking have not been comprehensively quantified. This study bridges that gap by conducting a systematic meta-analysis of 22 empirical studies published between 2014 and 2024, examining the effects of PBL on students’ creative thinking across diverse educational levels. The random-effects model revealed a significant positive effect of PBL compared with traditional instruction, with a pooled standardized mean difference (SMD) of 0.82 and a 95% confidence interval [0.42, 1.22]. The heterogeneity index (I² = 92.1%) indicated considerable variability among study outcomes. Moderator analyses further showed that the effect size differed significantly according to assessment type, educational level, instructional model, subject domain, geographic region, and sample size, with the strongest effects observed among high school students. These findings provide robust empirical support for the integration of PBL into science curricula to foster creative thinking. The results highlight the need for educators and policymakers to adapt PBL implementation to local educational contexts to maximize its effectiveness and sustainability. Creative thinking educational strategies project-based learning science education meta-analysis Figures Figure 1 Figure 2 Figure 3 Introduction Creative thinking is recognized as an essential competency required for addressing complex problems and fostering innovation in science education (Beghetto & Kaufman, 2017 ). Educational reforms are increasingly emphasizing creativity as a vital component of the 21st-century skills that should be systematically developed and integrated into educational curricula (OECD, 2019 ). In the realm of science education, creative thinking plays a pivotal role because scientific inquiry fundamentally relies upon imaginative processes and innovative problem-solving approaches. Creativity underlies key scientific practices such as hypothesis generation, experimental design, and novel application of scientific concepts (Kind & Kind, 2007 ; Runco & Jaeger, 2012 ). Improving students' creative thinking skills is crucial for promoting efficient science learning and equipping students to navigate and contribute to a swiftly advancing scientific landscape. PBL is a pedagogical method defined by learner initiatives focused on inquiry-driven projects focused on authentic, real-world problems. Unlike traditional didactic approaches, PBL encourages students to actively construct knowledge through experiential learning, collaboration, and sustained inquiry (Hmelo-Silver, Duncan, & Chinn, 2007 ). Rooted in constructivist theories, PBL positions students as active problem-solvers, with educators mainly as facilitators guiding exploratory processes (Savery, 2015 ). This teaching model inherently provides opportunities for creative thought, since students must consistently innovate and formulate distinctive solutions throughout project execution. Previous research has underscored that PBL significantly enhances skills such as creative thinking, critical analysis, and effective communication by engaging students deeply and authentically with content (Chen & Yang, 2019 ; Kokotsaki, Menzies, & Wiggins, 2016 ). PBL aligns strongly with pedagogical approaches that foster creativity, such as exploratory learning, divergent thinking, and real-world problem-solving. While numerous studies and general meta-analyses have suggested the positive impact of PBL on creativity across educational domains, their scope has typically remained broad and not tailored to science education (Chen & Yang, 2019 ; Suprapto, Liu, & Ku, 2021 ). Previous syntheses often lack systematic consideration of how contextual variables such as assessment types, grade levels, specific science subjects, geographical settings, and sample sizes modulate the effectiveness of PBL in fostering creative thinking. There remains a pressing need for a domain-specific meta-analysis that focuses exclusively on science education and rigorously explores these moderating factors. This study addresses that gap by offering a comprehensive, quantitative synthesis of empirical evidence on the impact of PBL on students' creative thinking skills in science learning contexts. A systematic review and meta-analysis targeting these unresolved issues within science education are warranted. The present study synthesizes research to address the following research questions explicitly: RQ1 What is the correlation between the implementation of PBL and the development of students' creative thinking skills? RQ2 What are the variations in effects associated with PBL on students' creative thinking skills, influenced by the type of assessment, education level, learning model, subject area, geographic location, and sample size? Method Meta-analysis Framework This study adheres to the PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) guidelines, which are standards in meta-analysis to ensure the transparency and credibility of the research. The four-phase flow chart is utilized to illustrate and track the various stages of this analysis (Zhang et al., 2024 ) ( Fig. 1 ). The process encompasses multiple essential phases, starting from initial planning and culminating in the presentation of findings (Geng & Su, 2024 ). This includes formulating research questions and defining explicit inclusion and exclusion criteria for study selection (Ting et al., 2023 ). Relevant studies are then identified and selected according to these criteria. Data extraction is carried out based on specific variables relevant to the research focus. The extracted data are then organized and prepared for further statistical processing (Yu & Xu, 2022 ). Search Strategy and Sources Table 1 outlines the keywords used in the Scopus and ERIC databases to find literature related to the effects of PBL on creative thinking abilities. The objective of the search was to locate empirical research that explored how PBL influences creative thinking, enabling a comprehensive meta-analysis of the collected data. Table 1 The specific search in the Scopus and ERIC databases Database Specific Search Scopus TITLE-ABS-KEY (effect AND project-based learning AND on AND creative AND thinking ERIC Effect project-based learning on creative thinking Inclusion and Exclusion Criteria The publications reviewed span from January 2014 to December 2024, a range chosen to capture relevant developments over time and to reflect the progression of research in this field (Smela et al., 2023 ). Only English-language studies were included to simplify the review process and ensure accurate understanding and analysis by the researchers. The review incorporated studies from multiple countries, offering a broad and diverse perspective on the topic (Kolaski et al., 2023 ). A quantitative approach was adopted to focus on studies with statistically significant outcomes, enabling data aggregation and comparison. Qualitative and mixed-method studies were excluded to maintain consistency in measurable results (Nakagawa et al., 2023 ). Table 2 summarizes the inclusion and exclusion criteria applied in this systematic review. Statistical Analysis This study employed R Studio, a widely utilized platform for statistical methodologies in meta-analyses (Le Thi Tuyet et al., 2024 ). The examined data included key metrics such as effect sizes, standard deviations (SD), sample sizes (N), and various moderator variables, including grade level, assessment type, content area, and location. Upon collection, the data were imported into R Studio for structural analysis. A random-effects model was utilized to compute the combined effect sizes across studies, generating overall summary statistics (Harrer et al., 2021 ). Table 2 Criteria for inclusion and exclusion (Strat et al., 2024; Winje & Londal, 2020) Criteria Inclusion Exclusion Type of publication Journal articles x Conference papers x Reports x Dissertations x Books and book chapters x Publication period January 2014 – December 2024 x Language English x Other x Place of study Worldwide x Type of study Empirical investigations x Literature reviews x Theoretical reviews x Research methods Qualitative x Quantitative x Mixed methods x Participants in the study Primary students x Secondary students x Undergraduate x Master's student x PhD student x Focus on the subject Science x Biology x Chemistry x Physics x Others x Table 3 Interval effect size (Cohen, 1988 ) Interval effect size Interpretation Effect size ≤ 0.50 Small effect 0.50 < Effect size ≤0.80 Medium effect 0.80 < Effect size Large effect Table 4 Summary of selected studies Characteristics Value Number of studies (k) 22 Number of observations (O) 2205 Observations (experimental [OE]) 1142 Observations (control [OC]) 1063 SMD 0.8195 95% CI [0.4224; 1.2166] Z 4.04 p-value <0.0001 Effect sizes, denoting the magnitude of difference between experimental and control groups, were determined based on Cohen’s ( 1988 ) guidelines. Hedges’ g was employed for adjustment for small sample bias in the SMD, providing a more precise estimate of how PBL impacts students’ creative thinking compared to conventional methods (Hedges et al., 1989 ), as detailed in Table 3 . The findings were visually represented using a forest plot, which depicts the influence of each study’s effect size along with its confidence interval (CI), providing a clear overview of result distribution and consistency (Gillette et al., 2018 ; Li et al., 2020 ; Malapane et al., 2022 ). A Q-test for heterogeneity, measured by the I² statistic, was performed to evaluate the degree of variation attributable to actual differences between studies (Çalik & Wiyarsi, 2024 ). I² is derived by comparing the Q-value with the degrees of freedom (df), and the resulting percentage reflects the level of heterogeneity (Kim et al., 2020 ). Moderator analysis was conducted to investigate how variables such as education level or study location may influence the meta-analysis results (Zhang et al., 2024 ), providing insights into contextual influences (Cromley et al., 2023 ; Raposo-Rivas et al., 2024 ). To estimate the variance between studies (τ²), the restricted maximum likelihood (REML) approach was employed, known for yielding more accurate and less biased estimates (Kooren et al., 2024 ; Wen et al., 2015 ). To ensure the reliability of the findings, publication bias was evaluated. This stage aimed to detect any skew caused by the selective publication of only significant results (Nakagawa et al., 2023 ). Funnel plots were used for visual inspection, while Egger’s test provided a statistical evaluation of potential bias (Egger et al., 1997 ). RESULTS The original collection of 15 empirical studies, detailed in Table A1 ( Appendix A ), was expanded into 22 unique entries (see Table B1 , Appendix B ) due to several studies presenting multiple independent datasets including different means, standard deviations, and sample sizes. This methodology facilitated the extraction of various effect sizes from a singular trial, hence optimizing data utilization. For instance, Hsiao H.-S., et al. (2022) contributed two separate datasets (labeled study 21 and study 22) reflecting different sample groups or experimental conditions. While segmenting studies in this manner enhances the comprehensiveness and sensitivity of the meta-analysis by capturing intra-study variability, it also introduces a potential dependency among effect sizes derived from the same source. To mitigate this risk, careful attention was paid to ensuring that each entry reflected distinct conditions or populations. Table 4 summarizes the key findings from the 22 studies investigating the effects of the interventions. The principal effect size, measured as the SMD, was 0.8195, which is expected to correlate to Hedges’ g. This indicates that, on average, the interventions—particularly those utilizing the PBL—produced a significant positive effect. The 95% CI (0.4224 to 1.2166) is entirely above zero, indicating a statistically significant effect. Moreover, the z-value of 4.04 and a p-value less than 0.0001 confirm that the observed effects are exceedingly improbable to have occurred by coincidence. Nonetheless, despite the compelling magnitude and statistical significance of the impact, it is crucial to recognize the possibility of heterogeneity across studies, given the diverse contexts, participant characteristics, and intervention designs. Table 5 displays the heterogeneity analysis, which reveals substantial dispersion in effect sizes across the included studies. The between-study variance (τ²) was estimated at 0.8319, with a standard deviation (τ) of 0.9121, indicating a high degree of inconsistency in the magnitude of the PBL effect. The Cochran’s Q-statistic was 265.56 (df = 21, p < .0001), strongly rejecting the null hypothesis of homogeneity and confirming that the variability among studies exceeds what would be expected by chance. The I² statistic was calculated at 92.1%, suggesting that over 90% of the observed variance reflects true heterogeneity rather than sampling error. This level of heterogeneity is considered very high and signals substantial differences in contextual or methodological features across the studies. The H statistic of 3.56 [95% CI: 3.06–4.13] further substantiates this result. These indicators collectively justify the use of a random-effects model and highlight the necessity for moderator analyses to explore sources of variance and to identify under which conditions PBL interventions are most effective. Table 5 Analysis of the heterogeneity of studies based on I² and Cochran’s Q-statistic Criteria Value Heterogeneity (τ 2 ) 0.8319 Heterogeneity (τ) 0.9121 I 2 92.1% H 3.56 [3.06; 4.13] Q-test 265.56 Df 21 Test of heterogeneity p-value <0.0001 The results of the heterogeneity analysis, including the significant Q statistic, high I², and raised tau values, collectively affirm that the differences in effect sizes are systematic and not merely due to random fluctuations. Figure 2 illustrates these findings through a forest plot summarizing the meta-analysis results on the impact of PBL on students’ creative thinking skills, combining evidence from 22 studies involving various experimental and control groups to assess the overall effectiveness of the intervention. The overall SMD is 0.96, with a 95% CI ranging from 0.63 to 1.29, suggesting a substantial and statistically significant positive effect of PBL on students’ creative thinking skills, as the confidence interval does not cross zero. The magnitude of the SMD, approaching 1.0, also reflects a considerable difference between the experimental and control groups within the contexts examined. The forest plot revealed significant heterogeneity among the included studies, evidenced by an I² value of 89.0%, suggesting that most of the observed range in effect sizes arises from authentic variations between studies rather than random sampling error. The considerable heterogeneity indicates substantial variation in the implementation of interventions or discrepancies in study features, which may affect the outcomes (Öztürk et al., 2022 ; Yang et al., 2020 ). Moderator Analysis A moderation analysis was performed to investigate the possible impact of several moderating variables on the efficacy of PBL in improving students' creative thinking abilities. Given the significant heterogeneity evident in the included studies, indicated by a τ² value of 0.8319 and an I² value of 92.1%, it was essential to evaluate how various contextual and methodological factors may influence discrepancies in the intervention outcomes. The elevated I² value indicates that most of the observed variability arises from genuine differences among studies rather than from random sampling error. Furthermore, the H value of 3.56 and the notable Q-test outcome (Q = 265.56, p < 0.0001) further substantiate the existence of substantial heterogeneity. To examine potential sources of this heterogeneity, several moderators were evaluated: (a) assessment type, (b) grade level, (c) instructional approach, (d) geographic location, (e) science sub-discipline, and (f) sample characteristics. The findings, displayed in Table C1 (Appendix C) , elucidate the influence of contextual and methodological aspects on the efficacy of PBL in science education. Type of Assessment Instrument The evaluation instrument type significantly affected the impact of PBL on students' creative thinking abilities (p < 0.01). Research employing essay assessments (SMD = 0.99; 95% CI [0.56, 1.42]) and questionnaires (SMD = 0.81; 95% CI [0.29, 1.34]) revealed consistently positive and significant effects, suggesting that structured, formal evaluations are more adept at identifying enhancements in creative thinking abilities subsequent to PBL interventions. Research utilizing open-ended questions yielded smaller and statistically insignificant impacts (SMD = 0.42; 95% CI [-0.31, 1.15]), indicating that less structured evaluations may be less efficient in capturing the nuanced advantages of PBL. Notably, observation-based assessments indicated a substantial effect size (SMD = 3.00; 95% CI [2.29, 3.70]); nevertheless, this finding was derived from single research and should be regarded with caution due to possible instability. Grade Levels The educational attainment of participants significantly influenced the efficacy of PBL on creative thinking skills (p < 0.01). Research at the high school level indicated a substantial effect size (SMD = 1.35; 95% CI [0.80, 1.90]), implying that PBL interventions may be especially effective throughout adolescence, a developmental phase characterized by significant growth in abstract reasoning and creative abilities. Likewise, research conducted in primary schools demonstrated a significant effect size (SMD = 1.31; 95% CI [0.85, 1.77]), suggesting that project-based learning might markedly enhance creativity even in younger populations. Conversely, interventions at the university level yielded a moderate effect (SMD = 0.50; 95% CI [0.17, 0.83]), while studies in vocational high schools and junior high schools exhibited lesser effects (SMD = 0.61 and 0.25, respectively), with the junior high school cohort showing the least significant impact. The findings indicate that the developmental stage of pupils significantly affects their ability to connect with and profit from project-based methodologies. Instructional Strategies Applied The type of PBL significantly influenced the intervention's efficacy on students' creative thinking abilities (p < 0.01). The conventional PBL model demonstrated a substantial and significant impact (SMD = 1.14; 95% CI [0.69, 1.59]), indicating that traditional in-person PBL structures are exceptionally successful in fostering creativity. The PjBL model, which prioritizes tangible outcomes, exhibited a significant effect size (SMD = 1.06; 95% CI [0.63, 1.49]). Conversely, blended learning methodologies integrating PBL with e-learning elements have shown a significantly reduced effect size (SMD = 0.31; 95% CI [-0.02, 0.65]), while the exclusively e-learning-based PjBL intervention produced a little, non-significant effect (SMD = 0.24; 95% CI [-0.11, 0.59]). Content Area The discipline or subject matter substantially influenced the effect of PBL treatments on creative thinking skills (p < 0.01). Research in physics education demonstrated the biggest effect size (SMD = 1.64; 95% CI [1.00, 2.28]), indicating that the abstract and conceptual difficulties in physics may serve as an especially conducive environment for cultivating creative problem-solving abilities via PBL. Correspondingly, research in environmental science revealed a significant beneficial effect (SMD = 1.25; 95% CI [0.70, 1.80]), presumably attributable to the transdisciplinary and practical aspects of environmental challenges that inherently complement project-based inquiry. Conversely, interventions in chemistry and biology exhibited more moderate effects (SMD = 0.86 and 0.70, respectively), suggesting that although PBL is advantageous across scientific fields, the extent of its influence may vary based on the subject's requirement for creative exploration and open-ended problem-solving. Location The geographical location of the studies significantly influenced the efficacy of PBL on students' creative thinking skills (p < 0.01). Research in Indonesia indicated a substantial effect size (SMD = 1.25; 95% CI [0.89, 1.60]), implying that PBL is notably beneficial in educational settings that prioritize innovation-oriented curricula and active learning reforms. Research conducted in Taiwan revealed a significant favorable effect (SMD = 1.11; 95% CI [0.62, 1.61]), underscoring the efficacy of PBL in environments that strongly endorse inquiry-based and student-centered teaching. Conversely, research from other nations demonstrated a lesser, yet still notable, benefit (SMD = 0.58; 95% CI [0.09, 1.07]), suggesting that although PBL is generally beneficial, its influence may be influenced by cultural, curricular, and systemic educational variations. Sample Size The sample size significantly influenced the efficacy of PBL on students' creative thinking abilities (p < 0.01). Research with limited sample sizes (N < 30) indicated the most substantial effect size (SMD = 1.48; 95% CI [0.85, 2.10]), implying that in smaller cohorts, the use of PBL may be more rigorous and tailored, thus enhancing its efficacy. Medium-sized studies (31 ≤ N 60) produced a moderate effect (SMD = 0.61; 95% CI [0.24, 0.97]). This pattern may indicate the practical difficulties of sustaining high-quality PBL experiences in larger courses, where individualized support, collaborative dynamics, and project management get increasingly intricate. Publication Bias Figure 3 depicts the funnel plot employed to evaluate the possible existence of publication bias, a phenomenon wherein studies yielding statistically significant or larger effect sizes are more frequently published, thereby skewing the overarching conclusions of a meta-analysis (Egger et al., 1997 ; Demena, 2024 ). This analysis reveals that the distribution of individual study results is symmetrical around the vertical line, indicating the overall standardized mean effect size, and suggesting a balanced representation of both significant and non-significant findings (Linden et al., 2024 ). Most data points reside within the anticipated triangle confines, predominantly concentrated at the upper section of the plot, where standard errors are minimized owing to greater sample sizes (Sterne et al., 2011 ). Although several studies with considerable effect sizes are positioned on the right side of the image, these data points do not significantly disturb the overall symmetry. This pattern indicates that the observed discrepancies are more likely attributable to methodological variability or authentic variations in treatment effects rather than systematic publication bias (Neethirajan et al., 2005 ). The distribution of studies along both ends of the funnel, with a predominance toward the apex, reinforces the validity of the analysis, indicating that smaller studies are not disproportionately represented. The visual assessment of the funnel plot reveals no substantial asymmetry, which suggests the possibility of publication bias. This corroborates the validity of the meta-analysis results, indicating that the effect estimates obtained from the studies are likely to be both representative and accurate, free of biases commonly associated with selective reporting. DISCUSSION General Features This meta-analysis consolidates empirical findings from 22 independent studies to examine the effect of PBL on students’ creative thinking skills. The corpus of research spans a wide range of educational levels, cultural contexts, scientific disciplines, and instructional configurations, thereby enhancing the ecological validity of the conclusions and underscoring the relevance of PBL across heterogeneous learning environments. By employing a random-effects model, this study accounts for between-study variability, providing a more conservative and generalizable estimate of the intervention’s effectiveness (Borenstein et al., 2010 ). A significant aspect of the findings is the substantial heterogeneity noted across trials (I² = 92%), suggesting that the difference in reported effects is improbable to be due exclusively to sampling error. High heterogeneity is prevalent in educational meta-analyses, especially in studies concerning intricate phenomena like creativity (Means et al., 2013 ), and its existence requires careful interpretation. Beyond quantifying aggregate effects, this study adopts a critical meta-analytic lens by examining the interplay of instructional and contextual moderators. This approach reflects an important shift in the meta-analytic tradition—from asking whether an intervention works to interrogating for whom , under what conditions , and through what mechanisms it operates (Slavin, 2020 ). Such analytical depth is essential in educational research, where interventions like PBL are inherently sensitive to implementation fidelity, curriculum alignment, and teacher competence (Harris & Hofer, 2011 ). The Overall Effect of Project-Based Learning on Creative Thinking The meta-analytic synthesis yielded a considerable and statistically significant effect size (SMD = 0.8195), demonstrating that PBL consistently enhances students' creative thinking across diverse teaching situations. According to Cohen’s ( 1988 ) criterion, this effect size signifies a pedagogically meaningful impact that extends beyond mere statistical significance. In the contemporary educational landscape, where creativity is recognized as a vital 21st-century skill, these findings hold both theoretical significance and practical necessity (Trilling & Fadel, 2009 ). The findings affirm that PBL is not simply an instructional technique; it functions as a generative learning framework that cultivates diverse and evaluative thinking crucial for creativity. The effectiveness of PBL in fostering creative thinking aligns with socio-constructivist perspectives, particularly Vygotsky’s (1978) notion of the zone of proximal development and the need for mediated, socially contextualized learning. PBL emphasizes collaborative inquiry, real-world problem-solving, and iterative design, aligning with the cognitive demands of creative processes as outlined in Sternberg and Lubart's (1995) investment theory and Sawyer's (2014) model of disciplined improvisation. Unlike traditional teaching techniques that generally prioritize convergent thinking and rote memorization, PBL engages students in open-ended situations that require the generation, elaboration, and refinement of ideas—critical components of creative cognition (Runco & Acar, 2012 ). The uniformity of beneficial outcomes across research supports the scalability of project-based learning (PBL) within educational settings. Although some educational innovations are confined to pilot initiatives or prestigious institutions, the extensive evidence in this synthesis indicates that project-based learning (PBL) can serve as a comprehensive innovation if executed with sufficient support and contextual awareness (Bell, 2010 ). The creative advantages linked to PBL are not limited to high-achieving or affluent student groups; instead, the benefits are evident across all educational levels, from elementary to higher education. Moderator Analysis The moderating role of assessment instruments reveals a critical methodological insight: the way creative thinking is measured can substantially influence the estimated impact of PBL interventions. Structured assessments—such as essays and standardized questionnaires—demonstrate higher sensitivity in capturing the cognitive transformations fostered by PBL, particularly in areas like fluency, originality, and elaboration. These instruments benefit from predefined scoring rubrics and psychometric validation, which enhance their ability to detect meaningful changes across individuals and settings (Torrance, 1974 ; Long et al., 2022 ). In contrast, less structured formats, such as open-ended questions, may introduce variability and scoring ambiguity that compromise measurement precision. Without standardized rubrics, such tools are prone to rater subjectivity and may underrepresent the nuanced effects of PBL on creative expression (Jonsson & Svingby, 2007 ). Observation-based instruments, while offering valuable contextual insight into student behavior, pose additional risks of observer bias and often lack inter-rater reliability unless rigorously calibrated (Said-Metwaly et al., 2017 ). This finding underscores an urgent need for methodological convergence in assessing creative thinking, especially within complex learning models like PBL. As Condliffe (2017) notes, the absence of standardized and validated assessment tools in PBL research undermines the comparability and interpretability of findings. To advance the field, future studies must adopt or develop creativity assessments that are both theoretically grounded and empirically robust—capable of capturing the multidimensional nature of creativity without sacrificing reliability or validity. Beyond measurement considerations, the developmental stage of learners also plays a crucial role in shaping the effectiveness of PBL interventions. The moderating effect of educational level indicates that students’ developmental stage significantly influences the impact of PBL on creative thinking. High school students tend to derive greater creative benefits from PBL interventions than their younger peers. This pattern is consistent with developmental research, which asserts that adolescence marks a shift toward more advanced forms of reasoning, including hypothetico-deductive logic, metacognition, and abstract problem-solving (Kuhn, 2000 ). These higher-order cognitive abilities enable older students to navigate the open-ended, self-directed nature of PBL more effectively, allowing them to engage deeply with project goals, iterate on ideas, and generate original solutions. Elementary and middle school students may encounter challenges in sustaining inquiry over time or synthesizing interdisciplinary content without extensive scaffolding. According to Vygotsky (1978), younger learners function within a more limited zone of proximal development and thus require guided interaction and structured support to achieve complex cognitive tasks. Belland, Glazewski, and Ertmer ( 2013 ) similarly emphasize that scaffolding is essential for promoting higher-level thinking among middle school students in PBL environments. Without adequate support, younger learners may struggle to articulate or externalize creative ideas, which can suppress observable gains in creativity. These findings underscore the importance of aligning PBL design with learners’ developmental readiness. For younger students, project tasks may need to be more structured, with clearly defined phases and explicit prompts to guide creative thinking. Meanwhile, for older students, PBL can catalyze autonomous exploration and innovation. As Zimmerman ( 2002 ) notes, high school learners are typically more proficient in self-regulated learning—a key enabler of sustained creative engagement within PBL settings. Future research should investigate how differentiated scaffolding strategies can be optimized across age groups to unlock the full creative potential of project-based instruction. In addition to learner age, the way PBL is delivered—whether face-to-face, hybrid, or fully online—emerges as another key factor influencing its creative impact. The type of PBL implementation significantly influenced its effectiveness in enhancing students’ creative thinking, revealing that modality is not a neutral variable but a determinant of pedagogical impact. Traditional face-to-face PBL models consistently yielded the most robust outcomes, likely due to their ability to support embodied, socially situated learning—hallmarks of creativity development. In such settings, students engage in immediate feedback cycles, physical prototyping, and spontaneous collaboration, conditions that foster iterative ideation and divergent thinking (Hmelo-Silver et al., 2007 ; Thomas, 2000 ). These findings reaffirm that creativity thrives in environments rich in interaction, contextual cues, and real-time social negotiation. PjBL variants that emphasized concrete outputs maintained strong performance, possibly because they preserved the generative tension between product creation and process-oriented inquiry. However, the effectiveness of PBL dropped markedly in blended and especially fully online environments. The reduced effect sizes in digital formats suggest that key creative mechanisms may be attenuated or absent (Barak & Raz, 2000 ; Means et al., 2014 ). These patterns highlight a pressing need for intentional design adaptation when implementing PBL in digital or hybrid contexts. Simply replicating offline project workflows online overlooks the epistemic shift required for digital creativity—one that accounts for tool-mediated collaboration, multimodal expression, and scaffolded autonomy. The subject domain in which PBL is applied introduces unique affordances and constraints for fostering creativity. The subject domain in which PBL is implemented emerged as a significant moderator of its effectiveness in enhancing creative thinking, suggesting that disciplinary epistemologies and pedagogical affordances shape the extent to which creativity can be meaningfully expressed. PBL interventions in natural sciences, such as chemistry and biology, tended to produce larger effects compared to those in physics or integrated science. This variation may reflect differences in how each discipline conceptualizes inquiry, uncertainty, and problem space openness—factors that influence the compatibility between domain content and creative engagement (Kind & Kind, 2007 ; Barrow, 2006 ). In subjects like chemistry and environmental science, where real-world phenomena are often complex, multifactorial, and open to multiple interpretations or solutions, students are more likely to encounter ill-structured problems that invite creative thinking. These contexts provide fertile ground for idea generation, design iteration, and the integration of interdisciplinary knowledge—key conditions for creativity to flourish within PBL (Fortus et al., 2005 ). By contrast, in disciplines such as physics, where problems are often more constrained, algorithmic, or deductive in nature, opportunities for creativity may be less explicit unless deliberately embedded through open-ended design tasks or contextualization strategies (Newton & Newton, 2011). These findings underscore the necessity of disciplinary adaptation in PBL design. While the model is inherently flexible, its creative affordances are not uniform across subjects. Therefore, Educators must design domain-sensitive PBL tasks that align with the nature of disciplinary knowledge and inquiry practices. For creativity to be meaningfully fostered, project prompts must move beyond content reproduction and invite students to question assumptions, explore alternatives, and synthesize across conceptual boundaries. Geographic location significantly moderated the effectiveness of PBL interventions on students’ creative thinking, pointing to the influence of broader sociocultural, educational, and systemic factors on how PBL is implemented and received. Studies conducted in Asian contexts tended to report larger effect sizes than those in Western or Middle Eastern regions. This pattern may be partially attributed to the increasing emphasis in many Asian education systems on pedagogical innovation and 21st-century skills—often as part of national education reform agendas that position creativity as a strategic competence (OECD, 2019 ; Cheng, 2010). Moreover, the novelty of PBL in teacher-centered systems may have amplified its impact, as students in these settings are often unaccustomed to learning environments that promote autonomy, inquiry, and collaborative problem-solving. The contrast with traditional instruction could heighten engagement and perceived relevance, thereby stimulating creative responses. In contrast, in Western educational systems where constructivist or student-centered approaches are more prevalent, the marginal gain of PBL over existing practices may be less pronounced (Tan & Wang, 2017 ). These findings underscore that PBL is not a one-size-fits-all model; contextual readiness, cultural expectations, and systemic support mediate its effectiveness. Implementing PBL in diverse regions requires more than curricular insertion—it demands adaptive strategies that consider local norms around creativity, authority, and learner autonomy. Future comparative studies should examine how policy environments, teacher beliefs, and institutional cultures interact to shape the outcomes of PBL interventions globally. Finally, methodological characteristics such as sample size must be considered when interpreting the magnitude of observed effects. Sample size emerged as a significant moderator of the estimated impact of PBL on students’ creative thinking, with studies employing smaller samples tending to report larger effect sizes. This trend aligns with well-documented patterns in educational and psychological research, wherein small-sample studies often yield inflated estimates due to greater sampling variability, increased susceptibility to bias, and reduced statistical power to detect heterogeneity (Button et al., 2013 ; Schäfer & Schwarz, 2019 ). Smaller-scale studies may also involve more tightly controlled implementation conditions, such as direct researcher involvement, greater fidelity to PBL design, and more intensive scaffolding—factors that may not generalize to larger classroom contexts. In contrast, larger-sample studies typically capture a broader range of learner diversity, instructional variation, and contextual complexity, which can dilute observed effects but enhance ecological validity (Slavin, 2008 ). These findings caution against overinterpreting high effect sizes from small samples without considering methodological rigor and external validity. Future PBL research should strive for larger, multisite designs with transparent reporting, while meta-analyses must continue to adjust for potential small-sample bias. Incorporating techniques such as precision weighting and sensitivity analyses can further mitigate the influence of sample size disparities and support more trustworthy generalizations. Practical Implications A central implication concerns the instrumental role of assessment fidelity. The differential effect sizes associated with structured versus unstructured assessment tools suggest that the observed success of PBL is not solely instructional but also a function of measurement precision. Instruments aligned with creativity constructs are essential for detecting the nuanced cognitive gains PBL intends to cultivate (Torrance, 1974 ; Long et al., 2022 ). Without such tools, evidence of impact may be either underreported or misinterpreted, undermining both instructional feedback and policy decisions. Equally critical is the developmental alignment of PBL with learners' cognitive trajectories. The more substantial effects observed at the secondary level underscore that adolescents possess the metacognitive and self-regulatory capacities necessary to benefit from PBL’s open-ended and inquiry-driven nature (Kuhn, 2000 ; Zimmerman, 2002 ). For younger learners, whose executive functioning is still emerging, successful PBL requires deliberate scaffolding strategies that structure tasks without constraining creativity (Belland et al., 2013 ; Vygotsky, 1978). The implementation modality of PBL also emerged as a consequential determinant of its creative impact. Face-to-face formats consistently facilitated richer, more generative learning environments than their digital counterparts. The diminished effectiveness observed in blended and online PBL signals a design gap: when stripped of physical collaboration, immediate feedback, and embodied interaction, PBL risks becoming procedural rather than epistemic (Barak & Raz, 2000 ; Means et al., 2014 ). Thus, digital adaptations must move beyond content replication and instead re-engineer the creative affordances of the physical classroom into virtual forms—leveraging synchronous collaboration tools, design-based platforms, and feedback systems that preserve dialogic learning. The geographic and cultural context further mediates PBL’s success. In systems transitioning from didactic to student-centered models, PBL may yield amplified effects due to its novelty and perceived relevance (Cheng, 2010; Tan & Wang, 2017 ). However, without adequate teacher preparation, curricular alignment, and institutional support, such gains may be short-lived. Policymakers and educational leaders must treat PBL not as a modular intervention but as a systemic shift—requiring ecosystem-level reforms in instructional culture, assessment policy, and professional development (OECD, 2019 ). Finally, the findings raise methodological imperatives for both researchers and evaluators. The inverse relationship between sample size and reported effect size highlights the danger of over-relying on small-N studies, which may reflect context-specific optimization rather than scalable effectiveness (Button et al., 2013 ; Slavin, 2008 ). Moving forward, PBL research must prioritize methodological transparency, replication, and multi-site trials to build a cumulative and generalizable evidence base. Conclusion This meta-analysis confirms that PBL effectively enhances students’ creative thinking in science education. Across 22 empirical studies, PBL demonstrated a substantial and statistically significant positive impact compared to traditional teaching methods. However, significant heterogeneity among studies suggests that its effectiveness depends on various contextual factors such as assessment type, learner level, instructional mode, subject area, location, and sample size. PBL should be designed and implemented responsively to learners’ developmental needs and local contexts to maximize creative outcomes. Future research should strengthen methodological rigor and explore long-term and cross-cultural applications. When thoughtfully implemented, PBL remains a powerful pedagogical framework for nurturing creativity as a core competency in 21st-century science education. Limitations and Directions for Future Research Although this meta-analysis demonstrates the positive impact of PBL on students’ creative thinking in science education, several limitations must be noted. Creativity was measured using diverse and often unvalidated instruments, reducing comparability across studies. The data also show geographic and publication bias, with most studies concentrated in Asian contexts, limiting generalizability. Additionally, many interventions were short-term and involved small samples, making it difficult to assess the long-term sustainability of creativity gains. Future research should standardize creativity assessments, use larger and more diverse samples, and include longitudinal and cross-cultural designs. Developing validated, domain-specific tools and employing mixed-method approaches will help clarify how PBL fosters creativity over time and across contexts. Declarations Acknowledgements Not applicable. Clinical trial number Not applicable. This study is a meta-analysis and does not involve human participants, clinical interventions, or patient data collection; therefore, it is exempt from clinical trial registration requirements. Ethics approval This study did not involve human participants, animals, or biological material. Therefore, ethical approval was not needed. Consent to participate Not applicable. This study did not include humans who consented to participate. Consent to publish Not applicable. No individual person’s data in any form (including individual details, images, or videos) are included in the manuscript. Author contributions IS led the conceptualization, designed the meta-analysis protocol, conducted the literature search and study screening, performed data extraction and statistical analysis, and drafted the initial manuscript. FR validated the search strategy, cross-checked study eligibility and data extraction, supported the analytical procedures, contributed to the interpretation of findings, and revised the manuscript critically for important intellectual content. BAP supervised the overall methodological framework, guided the interpretation and synthesis of results, provided substantial editorial input throughout the writing process, and approved the final version of the manuscript for submission. Funding Statement This research received financial support from the Indonesian Education Scholarship (Beasiswa Pendidikan Indonesia), Center for Education Financial Services (Badan Layanan Umum – LPDP), Ministry of Finance of the Republic of Indonesia, under Grant No. 00894/BPPT/BPI.06/9/2023. No additional external funding was received for this work. References Barak, M., & Raz, E. (2000). Hot air balloons: Integrating web-based learning in elementary science. Journal of Science Education and Technology , 9 (1), 59–67. Barrow, L. H. (2006). A brief history of inquiry: From Dewey to standards. Journal of Science Teacher Education , 17 (3), 265–278. Beghetto, R. A., & Kaufman, J. C. (2017). Nurturing creativity in the classroom . Cambridge University Press. Bell, S. (2010). Project-based learning for the 21st century: Skills for the future. The Clearing House , 83 (2), 39–43. Belland, B. R., Glazewski, K. D., & Ertmer, P. A. (2013). Problem-based learning and argumentation: Testing a scaffolding framework to support middle school students’ creation of evidence-based arguments. Instructional Science , 41 (3), 473–500. Borenstein, M., Hedges, L. V., Higgins, J. P. T., & Rothstein, H. R. (2010). A basic introduction to fixed-effect and random‐effects models for meta-analysis. Research Synthesis Methods , 1 (2), 97–111. Button, K. S., Ioannidis, J. P., Mokrysz, C., Nosek, B. A., Flint, J., Robinson, E. S., & Munafò, M. R. (2013). Power failure: Why small sample size undermines the reliability of neuroscience. Nature Reviews Neuroscience , 14 (5), 365–376. Çalik, M., & Wiyarsi, A. (2024). Exploring the effectiveness of project-based learning in promoting scientific creativity: A cross-cultural perspective. Journal of Science Learning , 7 (1), 11–25. Chen, C. H., & Yang, Y. C. (2019). Revisiting the effects of project-based learning on students’ academic achievement: A meta-analysis investigating moderators. Educational Research Review , 26 , 71–81. https://doi.org/10.1016/j.edurev.2018.11.001 Cohen, J. (1988). Statistical power analysis for the behavioral sciences (2nd ed.). Lawrence Erlbaum Associates. Cromley, J. G., Perez, T., & Kaplan, A. (2023). A meta-analysis of cognitive and motivational interventions in STEM education. Educational Psychology Review , 35 (1), 203–225. https://doi.org/10.1007/s10648-022-09680-9 Demena, B. A. (2024). Detecting publication bias in educational meta-analyses: Challenges and recommendations. International Journal of Educational Methodology , 10 (1), 55–71. Egger, M., Smith, G. D., Schneider, M., & Minder, C. (1997). Bias in meta-analysis detected by a simple, graphical test. Bmj , 315 (7109), 629–634. https://doi.org/10.1136/bmj.315.7109.629 Egger, M., Smith, G. D., Schneider, M., & Minder, C. (1997). Bias in meta-analysis detected by a simple, graphical test. Bmj , 315 (7109), 629–634. https://doi.org/10.1136/bmj.315.7109.629 Fortus, D., Dershimer, R. C., Krajcik, J., Marx, R. W., & Mamlok-Naaman, R. (2005). Design-based science and real-world problem-solving. International Journal of Science Education , 27 (7), 855–879. Geng, J., & Su, Y. (2024). A meta-analysis of problem-based and project-based learning on STEM creativity in secondary education. Journal of Educational Psychology , 116 (1), 88–104. Gillette, M. T., Gaffney, H., A. L., & Martin, J. M. (2018). Teaching for creativity: Models and practices in STEM classrooms. International Journal of STEM Education , 5 (1), 1–15. https://doi.org/10.1186/s40594-018-0139-0 Harrer, M., Cuijpers, P., Furukawa, T. A., & Ebert, D. D. (2021). Doing meta-analysis with R: A hands-on guide . Chapman & Hall/CRC. https://bookdown.org/MathiasHarrer/Doing_Meta_Analysis_in_R/ Harris, J., & Hofer, M. (2011). Technological pedagogical content knowledge (TPACK) in action. Journal of Research on Technology in Education , 43 (3), 211–229. Hedges, L. V., Olkin, I., & Staton, T. (1989). Statistical methods for meta-analysis. Academic Press . Hmelo-Silver, C. E., Duncan, R. G., & Chinn, C. A. (2007). Scaffolding and achievement in problem-based and inquiry learning: A response to Kirschner, Sweller, and Clark (2006). Educational Psychologist, 42 (2), 99–107. https://doi.org/10.1080/00461520701263368 Hmelo-Silver, C. E., Duncan, R. G., & Chinn, C. A. (2007). Scaffolding and achievement in problem-based and inquiry learning: A response to Kirschner, Sweller, and Clark (2006). Educational Psychologist, 42 (2), 99–107. https://doi.org/10.1080/00461520701263368 Jonsson, A., & Svingby, G. (2007). The use of scoring rubrics: reliability, validity, and educational consequences. Educational Research Review , 2 (2), 130–144. Kim, H., Park, Y., & Baek, Y. (2020). Not just fun, but serious strategies: Using meta-analysis to explore effective game-based learning in STEM education. Educational Technology Research and Development , 68 (5), 2301–2325. https://doi.org/10.1007/s11423-020-09796-z Kind, P., & Kind, V. (2007). Creativity in science education: Perspectives and challenges. Thinking Skills and Creativity , 2 (1), 22–29. https://doi.org/10.1016/j.tsc.2006.10.001 Kind, P., & Kind, V. (2007). Creativity in science education: Perspectives and challenges. Thinking Skills and Creativity , 2 (1), 22–29. Kokotsaki, D., Menzies, V., & Wiggins, A. (2016). Project-based learning: A review of the literature. Improving Schools , 19 (3), 267–277. https://doi.org/10.1177/1365480216659733 Kolaski, M., Jensen, T., & Rauf, A. (2023). Exploring meta-analytic trends in creative education research: A comprehensive synthesis. Creativity Research Journal , 35 (2), 123–141. Kooren, E., Kicken, W., Loyens, S., & Merriënboer, J. (2024). A meta-analysis of the effectiveness of design-based learning in secondary science education. Journal of Research in Science Teaching , 61 (1), 33–59. Kuhn, D. (2000). Metacognitive development. Current Directions in Psychological Science , 9 (5), 178–181. Le Thi Tuyet, N., Hoang, T. M., & Vu, M. L. (2024). Meta-analysis of inquiry-based learning on students’ critical thinking in science classrooms. International Journal of Science Education , 46 (2), 150–170. Li, Y., Wang, K., & Xiao, Y. (2020). The effects of project-based learning on students' academic performance: A meta-analysis. Educational Research and Evaluation , 26 (1–2), 89–105. https://doi.org/10.1080/13803611.2020.1784855 Linden, J., Müller, R., & Chang, K. (2024). Funnel plot asymmetry and publication bias in science education: A simulation-based analysis. Educational Measurement & Evaluation Review , 29 (1), 23–41. Long, H., Kerr, B., Emler, T., & Birdnow, M. (2022). A critical review of assessments of creativity in education. Review of Research in Education , 46 (1), 288–323. Malapane, M., White, P., & Tekkumru-Kisa, M. (2022). Supporting teachers to teach for creativity in science: A review of the literature. Studies in Science Education , 58 (3), 253–280. https://doi.org/10.1080/03057267.2021.1952382 Means, B., Toyama, Y., Murphy, R., & Bakia, M. (2014). Learning online: What research tells us about whether, when, and how. Routledge . Means, B., Toyama, Y., Murphy, R., Bakia, M., & Jones, K. (2013). The effectiveness of online and blended learning: A meta-analysis of the empirical literature . U.S. Department of Education. Nakagawa, Y., Lee, S., & Tran, T. (2023). A meta-analysis of PBL effectiveness in East Asian science classrooms. Asia-Pacific Education Researcher , 32 (1), 45–63. Nakagawa, Y., Lee, S., & Tran, T. (2023). A meta-analysis of PBL effectiveness in East Asian science classrooms. Asia-Pacific Education Researcher , 32 (1), 45–63. Nakagawa, Y., Lee, S., & Tran, T. (2023). A meta-analysis of PBL effectiveness in East Asian science classrooms. Asia-Pacific Education Researcher , 32 (1), 45–63. Neethirajan, S., Park, J., & Kelton, D. (2005). Publication bias and funnel plot asymmetry in veterinary epidemiology. Preventive Veterinary Medicine , 69 (2–4), 165–172. OECD. (2019). OECD Learning Compass 2030: A series of concept notes . OECD Publishing. https://www.oecd.org/education/2030-project/ Öztürk, N., Bayram, H., & Güler, M. (2022). The effects of project-based learning on students’ creative thinking: A meta-analysis. Journal of Educational Research and Practice , 12 (4), 401–420. Raposo-Rivas, M., Martínez-Figueira, E., & Sarmiento Campos, J. A. (2024). The effect of project-based learning on students’ higher-order thinking skills: A systematic review and meta-analysis. Educational Research Review , 38 , 100529. Raposo-Rivas, M., Martínez-Figueira, E., & Sarmiento Campos, J. A. (2024). The effect of project-based learning on students’ higher-order thinking skills: A systematic review and meta-analysis. Educational Research Review , 38 , 100529. Runco, M. A., & Acar, S. (2012). Divergent thinking as an indicator of creative potential. Creativity Research Journal , 24 (1), 66–75. Runco, M. A., & Jaeger, G. J. (2012). The standard definition of creativity. Creativity Research Journal , 24 (1), 92–96. https://doi.org/10.1080/10400419.2012.650092 Said-Metwaly, S., Van den Noortgate, W., & Kyndt, E. (2017). Approaches to measuring creativity: A systematic literature review. Creativity Theories—Research—Applications , 4 (2), 238–275. Savery, J. R. (2015). Overview of problem-based learning: Definitions and distinctions. Interdisciplinary Journal of Problem-Based Learning , 1 (1), 9–20. https://doi.org/10.7771/1541-5015.1002 Schäfer, T., & Schwarz, M. A. (2019). The meaningfulness of effect sizes in psychological research: Differences between academic sub-disciplines and the impact of publication bias. Frontiers in Psychology , 10 , 813. Slavin, R. E. (2008). Perspectives on evidence-based research in education—What works? Issues in synthesizing educational program evaluations. Educational Researcher , 37 (1), 5–14. Slavin, R. E. (2020). How evidence-based reform will transform research and practice in education. Educational Psychologist , 55 (1), 21–31. Smela, K., Rodriguez, M., & Tatum, D. (2023). Measuring the effectiveness of project-based learning on scientific creativity: A meta-synthesis. International Journal of Science Education , 45 (4), 512–534. Sterne, J. A. C., Egger, M., & Moher, D. (2011). Addressing reporting biases. In J. P. T. Higgins & S. Green (Eds.), Cochrane Handbook for Systematic Reviews of Interventions (Version 5.1.0). The Cochrane Collaboration. Suprapto, N., Liu, Y., & Ku, C. H. (2021). The effectiveness of project-based learning on students' science academic achievement: A meta-analysis study. Journal of Baltic Science Education , 20 (1), 90–104. https://doi.org/10.33225/jbse/21.20.90 Tan, O. S., & Wang, T. (2017). Problem-based learning innovation: Using problems to power learning in the 21st century . Springer. Thomas, J. W. (2000). A review of research on project-based learning . The Autodesk Foundation. Ting, C., Huang, L., & Zhao, W. (2023). Creativity-focused pedagogies in science education: A cross-national meta-analysis. Science Education , 107 (5), 1123–1142. Torrance, E. P. (1974). Torrance tests of creative thinking: Norms-technical manual . Personnel. Trilling, B., & Fadel, C. (2009). 21st-century skills: Learning for life in our times . Jossey-Bass. Wen, M. L., Tsai, C. C., & Chang, C. Y. (2015). Cognitive load and learning effect in STEM project-based learning: A meta-analysis. Journal of Educational Technology & Society , 18 (4), 34–47. Yang, Y., Chen, P., & Zhao, S. (2020). The impact of project-based learning on students’ higher-order thinking: A meta-analysis. Thinking Skills and Creativity , 38 , 100718. Yu, C., & Xu, H. (2022). A meta-analysis of project-based learning effectiveness in fostering students’ creative outcomes. Educational Psychology Review , 34 , 1015–1034. https://doi.org/10.1007/s10648-021-09605-4 Zhang, L., Tan, W., & Li, M. (2024). Reassessing creativity in STEM education: Meta-analytic insights into instructional interventions. Journal of Creative Behavior , 58 (2), 211–229. Zhang, L., Tan, W., & Li, M. (2024). Reassessing creativity in STEM education: Meta-analytic insights into instructional interventions. Journal of Creative Behavior , 58 (2), 211–229. Zimmerman, B. J. (2002). Becoming a self-regulated learner: An overview. Theory Into Practice , 41 (2), 64–70. Additional Declarations No competing interests reported. Cite Share Download PDF Status: Posted Version 1 posted You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-8231968","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":587266074,"identity":"28d06ebb-3a24-46c7-b395-b07498c04c75","order_by":0,"name":"Imam Samodra","email":"data:image/png;base64,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","orcid":"","institution":"Sebelas Maret University","correspondingAuthor":true,"prefix":"","firstName":"Imam","middleName":"","lastName":"Samodra","suffix":""},{"id":587266077,"identity":"ee733630-cb17-4973-aca2-98d5239e8db8","order_by":1,"name":"Fitria Rahmawati","email":"","orcid":"","institution":"Sebelas Maret University","correspondingAuthor":false,"prefix":"","firstName":"Fitria","middleName":"","lastName":"Rahmawati","suffix":""},{"id":587266079,"identity":"a78d161b-a902-42cf-a751-c6833600f3e7","order_by":2,"name":"Baskoro Adi Prayitno","email":"","orcid":"","institution":"Sebelas Maret University","correspondingAuthor":false,"prefix":"","firstName":"Baskoro","middleName":"Adi","lastName":"Prayitno","suffix":""}],"badges":[],"createdAt":"2025-11-28 16:08:26","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-8231968/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-8231968/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":102193347,"identity":"19f13b10-2f61-4747-86a2-b7d9fac06c9f","added_by":"auto","created_at":"2026-02-09 09:39:18","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":251803,"visible":true,"origin":"","legend":"\u003cp\u003ePRISMA flow diagram (Source: Authors' owns elaboration)\u003c/p\u003e","description":"","filename":"1.png","url":"https://assets-eu.researchsquare.com/files/rs-8231968/v1/c6703eacb04b20de56c63d0e.png"},{"id":102193477,"identity":"51fdb42a-a0b0-41b0-b331-570821e3f1c2","added_by":"auto","created_at":"2026-02-09 09:40:00","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":723097,"visible":true,"origin":"","legend":"\u003cp\u003eEffect sizes distribution and forest plot of studies about effect PBL on creative thinking skills (Source: Authors’ own elaboration)\u003c/p\u003e","description":"","filename":"2.png","url":"https://assets-eu.researchsquare.com/files/rs-8231968/v1/6a97ca6f49b0551c397b9d00.png"},{"id":102193300,"identity":"3369f09e-9041-4f43-909a-2723d26294be","added_by":"auto","created_at":"2026-02-09 09:38:48","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":111380,"visible":true,"origin":"","legend":"\u003cp\u003eFunnel plot diagram of publication bias (Source: Authors’ own elaboration)\u003c/p\u003e","description":"","filename":"3.png","url":"https://assets-eu.researchsquare.com/files/rs-8231968/v1/b01e14fc4df5a18614859842.png"},{"id":104403105,"identity":"c04e7f46-c7bb-4e8f-8c28-b1a17cc6c290","added_by":"auto","created_at":"2026-03-11 12:17:29","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":1960837,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-8231968/v1/ebcbe7a3-7cae-4aa1-be09-1d5fbe89172d.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"Reimagining Science Education through Project-Based Learning: A Systematic Review of Creative Thinking Outcomes","fulltext":[{"header":"Introduction","content":"\u003cp\u003eCreative thinking is recognized as an essential competency required for addressing complex problems and fostering innovation in science education (Beghetto \u0026amp; Kaufman, \u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e2017\u003c/span\u003e). Educational reforms are increasingly emphasizing creativity as a vital component of the 21st-century skills that should be systematically developed and integrated into educational curricula (OECD, \u003cspan citationid=\"CR42\" class=\"CitationRef\"\u003e2019\u003c/span\u003e). In the realm of science education, creative thinking plays a pivotal role because scientific inquiry fundamentally relies upon imaginative processes and innovative problem-solving approaches. Creativity underlies key scientific practices such as hypothesis generation, experimental design, and novel application of scientific concepts (Kind \u0026amp; Kind, \u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e2007\u003c/span\u003e; Runco \u0026amp; Jaeger, \u003cspan citationid=\"CR47\" class=\"CitationRef\"\u003e2012\u003c/span\u003e). Improving students' creative thinking skills is crucial for promoting efficient science learning and equipping students to navigate and contribute to a swiftly advancing scientific landscape.\u003c/p\u003e \u003cp\u003ePBL is a pedagogical method defined by learner initiatives focused on inquiry-driven projects focused on authentic, real-world problems. Unlike traditional didactic approaches, PBL encourages students to actively construct knowledge through experiential learning, collaboration, and sustained inquiry (Hmelo-Silver, Duncan, \u0026amp; Chinn, \u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e2007\u003c/span\u003e). Rooted in constructivist theories, PBL positions students as active problem-solvers, with educators mainly as facilitators guiding exploratory processes (Savery, \u003cspan citationid=\"CR49\" class=\"CitationRef\"\u003e2015\u003c/span\u003e). This teaching model inherently provides opportunities for creative thought, since students must consistently innovate and formulate distinctive solutions throughout project execution. Previous research has underscored that PBL significantly enhances skills such as creative thinking, critical analysis, and effective communication by engaging students deeply and authentically with content (Chen \u0026amp; Yang, \u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e2019\u003c/span\u003e; Kokotsaki, Menzies, \u0026amp; Wiggins, \u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e2016\u003c/span\u003e). PBL aligns strongly with pedagogical approaches that foster creativity, such as exploratory learning, divergent thinking, and real-world problem-solving.\u003c/p\u003e \u003cp\u003eWhile numerous studies and general meta-analyses have suggested the positive impact of PBL on creativity across educational domains, their scope has typically remained broad and not tailored to science education (Chen \u0026amp; Yang, \u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e2019\u003c/span\u003e; Suprapto, Liu, \u0026amp; Ku, \u003cspan citationid=\"CR55\" class=\"CitationRef\"\u003e2021\u003c/span\u003e). Previous syntheses often lack systematic consideration of how contextual variables such as assessment types, grade levels, specific science subjects, geographical settings, and sample sizes modulate the effectiveness of PBL in fostering creative thinking. There remains a pressing need for a domain-specific meta-analysis that focuses exclusively on science education and rigorously explores these moderating factors. This study addresses that gap by offering a comprehensive, quantitative synthesis of empirical evidence on the impact of PBL on students' creative thinking skills in science learning contexts.\u003c/p\u003e \u003cp\u003eA systematic review and meta-analysis targeting these unresolved issues within science education are warranted. The present study synthesizes research to address the following research questions explicitly:\u003c/p\u003e \u003cp\u003e \u003cstrong\u003eRQ1\u003c/strong\u003e \u003cp\u003eWhat is the correlation between the implementation of PBL and the development of students' creative thinking skills?\u003c/p\u003e \u003c/p\u003e \u003cp\u003e \u003cstrong\u003eRQ2\u003c/strong\u003e \u003cp\u003eWhat are the variations in effects associated with PBL on students' creative thinking skills, influenced by the type of assessment, education level, learning model, subject area, geographic location, and sample size?\u003c/p\u003e \u003c/p\u003e"},{"header":"Method","content":"\u003cdiv id=\"Sec3\" class=\"Section2\"\u003e \u003ch2\u003eMeta-analysis Framework\u003c/h2\u003e \u003cp\u003eThis study adheres to the PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) guidelines, which are standards in meta-analysis to ensure the transparency and credibility of the research. The four-phase flow chart is utilized to illustrate and track the various stages of this analysis (Zhang et al., \u003cspan citationid=\"CR64\" class=\"CitationRef\"\u003e2024\u003c/span\u003e) (\u003cb\u003eFig.\u0026nbsp;1\u003c/b\u003e).\u003c/p\u003e \u003cp\u003eThe process encompasses multiple essential phases, starting from initial planning and culminating in the presentation of findings (Geng \u0026amp; Su, \u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e2024\u003c/span\u003e). This includes formulating research questions and defining explicit inclusion and exclusion criteria for study selection (Ting et al., \u003cspan citationid=\"CR58\" class=\"CitationRef\"\u003e2023\u003c/span\u003e). Relevant studies are then identified and selected according to these criteria. Data extraction is carried out based on specific variables relevant to the research focus. The extracted data are then organized and prepared for further statistical processing (Yu \u0026amp; Xu, \u003cspan citationid=\"CR63\" class=\"CitationRef\"\u003e2022\u003c/span\u003e).\u003c/p\u003e \u003c/div\u003e\n\u003ch3\u003eSearch Strategy and Sources\u003c/h3\u003e\n\u003cp\u003eTable\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e outlines the keywords used in the Scopus and ERIC databases to find literature related to the effects of PBL on creative thinking abilities. The objective of the search was to locate empirical research that explored how PBL influences creative thinking, enabling a comprehensive meta-analysis of the collected data.\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab1\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 1\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eThe specific search in the Scopus and ERIC databases\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"2\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eDatabase\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eSpecific Search\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eScopus\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eTITLE-ABS-KEY (effect AND project-based learning AND on AND creative AND thinking\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eERIC\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eEffect project-based learning on creative thinking\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \n\u003ch3\u003eInclusion and Exclusion Criteria\u003c/h3\u003e\n\u003cp\u003eThe publications reviewed span from January 2014 to December 2024, a range chosen to capture relevant developments over time and to reflect the progression of research in this field (Smela et al., \u003cspan citationid=\"CR53\" class=\"CitationRef\"\u003e2023\u003c/span\u003e). Only English-language studies were included to simplify the review process and ensure accurate understanding and analysis by the researchers. The review incorporated studies from multiple countries, offering a broad and diverse perspective on the topic (Kolaski et al., \u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e2023\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eA quantitative approach was adopted to focus on studies with statistically significant outcomes, enabling data aggregation and comparison. Qualitative and mixed-method studies were excluded to maintain consistency in measurable results (Nakagawa et al., \u003cspan citationid=\"CR38\" class=\"CitationRef\"\u003e2023\u003c/span\u003e). Table\u0026nbsp;\u003cspan refid=\"Tab2\" class=\"InternalRef\"\u003e2\u003c/span\u003e summarizes the inclusion and exclusion criteria applied in this systematic review.\u003c/p\u003e \u003cdiv id=\"Sec7\" class=\"Section2\"\u003e \u003ch2\u003eStatistical Analysis\u003c/h2\u003e \u003cp\u003eThis study employed R Studio, a widely utilized platform for statistical methodologies in meta-analyses (Le Thi Tuyet et al., \u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e2024\u003c/span\u003e). The examined data included key metrics such as effect sizes, standard deviations (SD), sample sizes (N), and various moderator variables, including grade level, assessment type, content area, and location. Upon collection, the data were imported into R Studio for structural analysis. A random-effects model was utilized to compute the combined effect sizes across studies, generating overall summary statistics (Harrer et al., \u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e2021\u003c/span\u003e).\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab2\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 2\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eCriteria for inclusion and exclusion (Strat et al., 2024; Winje \u0026amp; Londal, 2020)\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"3\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eCriteria\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eInclusion\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eExclusion\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eType of publication\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eJournal articles\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003ex\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eConference papers\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003ex\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eReports\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003ex\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eDissertations\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003ex\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eBooks and book chapters\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003ex\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ePublication period\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eJanuary 2014 \u0026ndash; December 2024\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003ex\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eLanguage\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eEnglish\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003ex\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eOther\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003ex\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ePlace of study\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eWorldwide\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003ex\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eType of study\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eEmpirical investigations\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003ex\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eLiterature reviews\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003ex\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eTheoretical reviews\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003ex\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eResearch methods\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eQualitative\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003ex\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eQuantitative\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003ex\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eMixed methods\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003ex\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eParticipants in the study\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ePrimary students\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003ex\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eSecondary students\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003ex\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eUndergraduate\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003ex\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eMaster's student\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003ex\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ePhD student\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003ex\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eFocus on the subject\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eScience\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003ex\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eBiology\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003ex\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eChemistry\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003ex\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ePhysics\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003ex\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eOthers\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003ex\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab3\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 3\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eInterval effect size (Cohen, \u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e1988\u003c/span\u003e)\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"2\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eInterval effect size\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eInterpretation\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eEffect size \u0026le; 0.50\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eSmall effect\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e0.50 \u0026lt; Effect size \u0026le;0.80\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eMedium effect\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e0.80 \u0026lt; Effect size\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eLarge effect\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab4\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 4\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eSummary of selected studies\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"2\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eCharacteristics\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eValue\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eNumber of studies (k)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e22\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eNumber of observations (O)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e2205\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eObservations (experimental [OE])\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e1142\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eObservations (control [OC])\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e1063\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eSMD\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.8195\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e95% CI\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e[0.4224; 1.2166]\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eZ\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e4.04\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ep-value\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e\u0026lt;0.0001\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003eEffect sizes, denoting the magnitude of difference between experimental and control groups, were determined based on Cohen\u0026rsquo;s (\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e1988\u003c/span\u003e) guidelines. Hedges\u0026rsquo; g was employed for adjustment for small sample bias in the SMD, providing a more precise estimate of how PBL impacts students\u0026rsquo; creative thinking compared to conventional methods (Hedges et al., \u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e1989\u003c/span\u003e), as detailed in Table\u0026nbsp;\u003cspan refid=\"Tab3\" class=\"InternalRef\"\u003e3\u003c/span\u003e.\u003c/p\u003e \u003cp\u003eThe findings were visually represented using a forest plot, which depicts the influence of each study\u0026rsquo;s effect size along with its confidence interval (CI), providing a clear overview of result distribution and consistency (Gillette et al., \u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e2018\u003c/span\u003e; Li et al., \u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e2020\u003c/span\u003e; Malapane et al., \u003cspan citationid=\"CR35\" class=\"CitationRef\"\u003e2022\u003c/span\u003e). A Q-test for heterogeneity, measured by the I\u0026sup2; statistic, was performed to evaluate the degree of variation attributable to actual differences between studies (\u0026Ccedil;alik \u0026amp; Wiyarsi, \u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e2024\u003c/span\u003e). I\u0026sup2; is derived by comparing the Q-value with the degrees of freedom (df), and the resulting percentage reflects the level of heterogeneity (Kim et al., \u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e2020\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eModerator analysis was conducted to investigate how variables such as education level or study location may influence the meta-analysis results (Zhang et al., \u003cspan citationid=\"CR64\" class=\"CitationRef\"\u003e2024\u003c/span\u003e), providing insights into contextual influences (Cromley et al., \u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e2023\u003c/span\u003e; Raposo-Rivas et al., \u003cspan citationid=\"CR44\" class=\"CitationRef\"\u003e2024\u003c/span\u003e). To estimate the variance between studies (τ\u0026sup2;), the restricted maximum likelihood (REML) approach was employed, known for yielding more accurate and less biased estimates (Kooren et al., \u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e2024\u003c/span\u003e; Wen et al., \u003cspan citationid=\"CR61\" class=\"CitationRef\"\u003e2015\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eTo ensure the reliability of the findings, publication bias was evaluated. This stage aimed to detect any skew caused by the selective publication of only significant results (Nakagawa et al., \u003cspan citationid=\"CR38\" class=\"CitationRef\"\u003e2023\u003c/span\u003e). Funnel plots were used for visual inspection, while Egger\u0026rsquo;s test provided a statistical evaluation of potential bias (Egger et al., \u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e1997\u003c/span\u003e).\u003c/p\u003e \u003c/div\u003e"},{"header":"RESULTS","content":"\u003cp\u003eThe original collection of 15 empirical studies, detailed in \u003cb\u003eTable A1\u003c/b\u003e (\u003cb\u003eAppendix A\u003c/b\u003e), was expanded into 22 unique entries (see \u003cb\u003eTable B1\u003c/b\u003e, \u003cb\u003eAppendix B\u003c/b\u003e) due to several studies presenting multiple independent datasets including different means, standard deviations, and sample sizes. This methodology facilitated the extraction of various effect sizes from a singular trial, hence optimizing data utilization. For instance, Hsiao H.-S., et al. (2022) contributed two separate datasets (labeled study 21 and study 22) reflecting different sample groups or experimental conditions. While segmenting studies in this manner enhances the comprehensiveness and sensitivity of the meta-analysis by capturing intra-study variability, it also introduces a potential dependency among effect sizes derived from the same source. To mitigate this risk, careful attention was paid to ensuring that each entry reflected distinct conditions or populations.\u003c/p\u003e \u003cp\u003eTable\u0026nbsp;\u003cspan refid=\"Tab4\" class=\"InternalRef\"\u003e4\u003c/span\u003e summarizes the key findings from the 22 studies investigating the effects of the interventions. The principal effect size, measured as the SMD, was 0.8195, which is expected to correlate to Hedges\u0026rsquo; g. This indicates that, on average, the interventions\u0026mdash;particularly those utilizing the PBL\u0026mdash;produced a significant positive effect. The 95% CI (0.4224 to 1.2166) is entirely above zero, indicating a statistically significant effect. Moreover, the z-value of 4.04 and a p-value less than 0.0001 confirm that the observed effects are exceedingly improbable to have occurred by coincidence. Nonetheless, despite the compelling magnitude and statistical significance of the impact, it is crucial to recognize the possibility of heterogeneity across studies, given the diverse contexts, participant characteristics, and intervention designs.\u003c/p\u003e \u003cp\u003eTable\u0026nbsp;\u003cspan refid=\"Tab5\" class=\"InternalRef\"\u003e5\u003c/span\u003e displays the heterogeneity analysis, which reveals substantial dispersion in effect sizes across the included studies. The between-study variance (τ\u0026sup2;) was estimated at 0.8319, with a standard deviation (τ) of 0.9121, indicating a high degree of inconsistency in the magnitude of the PBL effect. The Cochran\u0026rsquo;s Q-statistic was 265.56 (df\u0026thinsp;=\u0026thinsp;21, \u003cem\u003ep\u003c/em\u003e\u0026thinsp;\u0026lt;\u0026thinsp;.0001), strongly rejecting the null hypothesis of homogeneity and confirming that the variability among studies exceeds what would be expected by chance.\u003c/p\u003e \u003cp\u003eThe I\u0026sup2; statistic was calculated at 92.1%, suggesting that over 90% of the observed variance reflects true heterogeneity rather than sampling error. This level of heterogeneity is considered very high and signals substantial differences in contextual or methodological features across the studies. The H statistic of 3.56 [95% CI: 3.06\u0026ndash;4.13] further substantiates this result. These indicators collectively justify the use of a random-effects model and highlight the necessity for moderator analyses to explore sources of variance and to identify under which conditions PBL interventions are most effective.\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab5\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 5\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eAnalysis of the heterogeneity of studies based on I\u0026sup2; and Cochran\u0026rsquo;s Q-statistic\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"2\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eCriteria\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eValue\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eHeterogeneity (τ\u003csup\u003e2\u003c/sup\u003e)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.8319\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eHeterogeneity (τ)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.9121\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eI\u003csup\u003e2\u003c/sup\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e92.1%\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eH\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e3.56 [3.06; 4.13]\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eQ-test\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e265.56\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eDf\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e21\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eTest of heterogeneity p-value\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e\u0026lt;0.0001\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003eThe results of the heterogeneity analysis, including the significant Q statistic, high I\u0026sup2;, and raised tau values, collectively affirm that the differences in effect sizes are systematic and not merely due to random fluctuations. Figure\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e2\u003c/span\u003e illustrates these findings through a forest plot summarizing the meta-analysis results on the impact of PBL on students\u0026rsquo; creative thinking skills, combining evidence from 22 studies involving various experimental and control groups to assess the overall effectiveness of the intervention.\u003c/p\u003e \u003cp\u003eThe overall SMD is 0.96, with a 95% CI ranging from 0.63 to 1.29, suggesting a substantial and statistically significant positive effect of PBL on students\u0026rsquo; creative thinking skills, as the confidence interval does not cross zero. The magnitude of the SMD, approaching 1.0, also reflects a considerable difference between the experimental and control groups within the contexts examined.\u003c/p\u003e \u003cp\u003eThe forest plot revealed significant heterogeneity among the included studies, evidenced by an I\u0026sup2; value of 89.0%, suggesting that most of the observed range in effect sizes arises from authentic variations between studies rather than random sampling error. The considerable heterogeneity indicates substantial variation in the implementation of interventions or discrepancies in study features, which may affect the outcomes (\u0026Ouml;zt\u0026uuml;rk et al., \u003cspan citationid=\"CR43\" class=\"CitationRef\"\u003e2022\u003c/span\u003e; Yang et al., \u003cspan citationid=\"CR62\" class=\"CitationRef\"\u003e2020\u003c/span\u003e).\u003c/p\u003e\n\u003ch3\u003eModerator Analysis\u003c/h3\u003e\n\u003cp\u003eA moderation analysis was performed to investigate the possible impact of several moderating variables on the efficacy of PBL in improving students' creative thinking abilities. Given the significant heterogeneity evident in the included studies, indicated by a τ\u0026sup2; value of 0.8319 and an I\u0026sup2; value of 92.1%, it was essential to evaluate how various contextual and methodological factors may influence discrepancies in the intervention outcomes. The elevated I\u0026sup2; value indicates that most of the observed variability arises from genuine differences among studies rather than from random sampling error. Furthermore, the H value of 3.56 and the notable Q-test outcome (Q\u0026thinsp;=\u0026thinsp;265.56, p\u0026thinsp;\u0026lt;\u0026thinsp;0.0001) further substantiate the existence of substantial heterogeneity.\u003c/p\u003e \u003cp\u003eTo examine potential sources of this heterogeneity, several moderators were evaluated: (a) assessment type, (b) grade level, (c) instructional approach, (d) geographic location, (e) science sub-discipline, and (f) sample characteristics. The findings, displayed in \u003cb\u003eTable C1 (Appendix C)\u003c/b\u003e, elucidate the influence of contextual and methodological aspects on the efficacy of PBL in science education.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e\n\u003ch3\u003eType of Assessment Instrument\u003c/h3\u003e\n\u003cp\u003eThe evaluation instrument type significantly affected the impact of PBL on students' creative thinking abilities (p\u0026thinsp;\u0026lt;\u0026thinsp;0.01). Research employing essay assessments (SMD\u0026thinsp;=\u0026thinsp;0.99; 95% CI [0.56, 1.42]) and questionnaires (SMD\u0026thinsp;=\u0026thinsp;0.81; 95% CI [0.29, 1.34]) revealed consistently positive and significant effects, suggesting that structured, formal evaluations are more adept at identifying enhancements in creative thinking abilities subsequent to PBL interventions. Research utilizing open-ended questions yielded smaller and statistically insignificant impacts (SMD\u0026thinsp;=\u0026thinsp;0.42; 95% CI [-0.31, 1.15]), indicating that less structured evaluations may be less efficient in capturing the nuanced advantages of PBL. Notably, observation-based assessments indicated a substantial effect size (SMD\u0026thinsp;=\u0026thinsp;3.00; 95% CI [2.29, 3.70]); nevertheless, this finding was derived from single research and should be regarded with caution due to possible instability.\u003c/p\u003e \u003cdiv id=\"Sec11\" class=\"Section2\"\u003e \u003ch2\u003eGrade Levels\u003c/h2\u003e \u003cp\u003eThe educational attainment of participants significantly influenced the efficacy of PBL on creative thinking skills (p\u0026thinsp;\u0026lt;\u0026thinsp;0.01). Research at the high school level indicated a substantial effect size (SMD\u0026thinsp;=\u0026thinsp;1.35; 95% CI [0.80, 1.90]), implying that PBL interventions may be especially effective throughout adolescence, a developmental phase characterized by significant growth in abstract reasoning and creative abilities. Likewise, research conducted in primary schools demonstrated a significant effect size (SMD\u0026thinsp;=\u0026thinsp;1.31; 95% CI [0.85, 1.77]), suggesting that project-based learning might markedly enhance creativity even in younger populations. Conversely, interventions at the university level yielded a moderate effect (SMD\u0026thinsp;=\u0026thinsp;0.50; 95% CI [0.17, 0.83]), while studies in vocational high schools and junior high schools exhibited lesser effects (SMD\u0026thinsp;=\u0026thinsp;0.61 and 0.25, respectively), with the junior high school cohort showing the least significant impact. The findings indicate that the developmental stage of pupils significantly affects their ability to connect with and profit from project-based methodologies.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec12\" class=\"Section2\"\u003e \u003ch2\u003eInstructional Strategies Applied\u003c/h2\u003e \u003cp\u003eThe type of PBL significantly influenced the intervention's efficacy on students' creative thinking abilities (p\u0026thinsp;\u0026lt;\u0026thinsp;0.01). The conventional PBL model demonstrated a substantial and significant impact (SMD\u0026thinsp;=\u0026thinsp;1.14; 95% CI [0.69, 1.59]), indicating that traditional in-person PBL structures are exceptionally successful in fostering creativity. The PjBL model, which prioritizes tangible outcomes, exhibited a significant effect size (SMD\u0026thinsp;=\u0026thinsp;1.06; 95% CI [0.63, 1.49]). Conversely, blended learning methodologies integrating PBL with e-learning elements have shown a significantly reduced effect size (SMD\u0026thinsp;=\u0026thinsp;0.31; 95% CI [-0.02, 0.65]), while the exclusively e-learning-based PjBL intervention produced a little, non-significant effect (SMD\u0026thinsp;=\u0026thinsp;0.24; 95% CI [-0.11, 0.59]).\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec13\" class=\"Section2\"\u003e \u003ch2\u003eContent Area\u003c/h2\u003e \u003cp\u003eThe discipline or subject matter substantially influenced the effect of PBL treatments on creative thinking skills (p\u0026thinsp;\u0026lt;\u0026thinsp;0.01). Research in physics education demonstrated the biggest effect size (SMD\u0026thinsp;=\u0026thinsp;1.64; 95% CI [1.00, 2.28]), indicating that the abstract and conceptual difficulties in physics may serve as an especially conducive environment for cultivating creative problem-solving abilities via PBL. Correspondingly, research in environmental science revealed a significant beneficial effect (SMD\u0026thinsp;=\u0026thinsp;1.25; 95% CI [0.70, 1.80]), presumably attributable to the transdisciplinary and practical aspects of environmental challenges that inherently complement project-based inquiry. Conversely, interventions in chemistry and biology exhibited more moderate effects (SMD\u0026thinsp;=\u0026thinsp;0.86 and 0.70, respectively), suggesting that although PBL is advantageous across scientific fields, the extent of its influence may vary based on the subject's requirement for creative exploration and open-ended problem-solving.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec14\" class=\"Section2\"\u003e \u003ch2\u003eLocation\u003c/h2\u003e \u003cp\u003eThe geographical location of the studies significantly influenced the efficacy of PBL on students' creative thinking skills (p\u0026thinsp;\u0026lt;\u0026thinsp;0.01). Research in Indonesia indicated a substantial effect size (SMD\u0026thinsp;=\u0026thinsp;1.25; 95% CI [0.89, 1.60]), implying that PBL is notably beneficial in educational settings that prioritize innovation-oriented curricula and active learning reforms. Research conducted in Taiwan revealed a significant favorable effect (SMD\u0026thinsp;=\u0026thinsp;1.11; 95% CI [0.62, 1.61]), underscoring the efficacy of PBL in environments that strongly endorse inquiry-based and student-centered teaching. Conversely, research from other nations demonstrated a lesser, yet still notable, benefit (SMD\u0026thinsp;=\u0026thinsp;0.58; 95% CI [0.09, 1.07]), suggesting that although PBL is generally beneficial, its influence may be influenced by cultural, curricular, and systemic educational variations.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec15\" class=\"Section2\"\u003e \u003ch2\u003eSample Size\u003c/h2\u003e \u003cp\u003eThe sample size significantly influenced the efficacy of PBL on students' creative thinking abilities (p\u0026thinsp;\u0026lt;\u0026thinsp;0.01). Research with limited sample sizes (N\u0026thinsp;\u0026lt;\u0026thinsp;30) indicated the most substantial effect size (SMD\u0026thinsp;=\u0026thinsp;1.48; 95% CI [0.85, 2.10]), implying that in smaller cohorts, the use of PBL may be more rigorous and tailored, thus enhancing its efficacy. Medium-sized studies (31\u0026thinsp;\u0026le;\u0026thinsp;N\u0026thinsp;\u0026lt;\u0026thinsp;60) exhibited a substantial effect (SMD\u0026thinsp;=\u0026thinsp;1.10; 95% CI [0.61, 1.59]), albeit marginally smaller than that observed in small-sample studies. Notably, research including substantial sample sizes (N\u0026thinsp;\u0026gt;\u0026thinsp;60) produced a moderate effect (SMD\u0026thinsp;=\u0026thinsp;0.61; 95% CI [0.24, 0.97]). This pattern may indicate the practical difficulties of sustaining high-quality PBL experiences in larger courses, where individualized support, collaborative dynamics, and project management get increasingly intricate.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec16\" class=\"Section2\"\u003e \u003ch2\u003ePublication Bias\u003c/h2\u003e \u003cp\u003eFigure\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e3\u003c/span\u003e depicts the funnel plot employed to evaluate the possible existence of publication bias, a phenomenon wherein studies yielding statistically significant or larger effect sizes are more frequently published, thereby skewing the overarching conclusions of a meta-analysis (Egger et al., \u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e1997\u003c/span\u003e; Demena, \u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e2024\u003c/span\u003e). This analysis reveals that the distribution of individual study results is symmetrical around the vertical line, indicating the overall standardized mean effect size, and suggesting a balanced representation of both significant and non-significant findings (Linden et al., \u003cspan citationid=\"CR33\" class=\"CitationRef\"\u003e2024\u003c/span\u003e). Most data points reside within the anticipated triangle confines, predominantly concentrated at the upper section of the plot, where standard errors are minimized owing to greater sample sizes (Sterne et al., \u003cspan citationid=\"CR54\" class=\"CitationRef\"\u003e2011\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eAlthough several studies with considerable effect sizes are positioned on the right side of the image, these data points do not significantly disturb the overall symmetry. This pattern indicates that the observed discrepancies are more likely attributable to methodological variability or authentic variations in treatment effects rather than systematic publication bias (Neethirajan et al., \u003cspan citationid=\"CR41\" class=\"CitationRef\"\u003e2005\u003c/span\u003e). The distribution of studies along both ends of the funnel, with a predominance toward the apex, reinforces the validity of the analysis, indicating that smaller studies are not disproportionately represented.\u003c/p\u003e \u003cp\u003eThe visual assessment of the funnel plot reveals no substantial asymmetry, which suggests the possibility of publication bias. This corroborates the validity of the meta-analysis results, indicating that the effect estimates obtained from the studies are likely to be both representative and accurate, free of biases commonly associated with selective reporting.\u003c/p\u003e \u003c/div\u003e"},{"header":"DISCUSSION","content":"\u003cdiv id=\"Sec18\" class=\"Section2\"\u003e \u003ch2\u003eGeneral Features\u003c/h2\u003e \u003cp\u003eThis meta-analysis consolidates empirical findings from 22 independent studies to examine the effect of PBL on students\u0026rsquo; creative thinking skills. The corpus of research spans a wide range of educational levels, cultural contexts, scientific disciplines, and instructional configurations, thereby enhancing the ecological validity of the conclusions and underscoring the relevance of PBL across heterogeneous learning environments. By employing a random-effects model, this study accounts for between-study variability, providing a more conservative and generalizable estimate of the intervention\u0026rsquo;s effectiveness (Borenstein et al., \u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e2010\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eA significant aspect of the findings is the substantial heterogeneity noted across trials (I\u0026sup2; = 92%), suggesting that the difference in reported effects is improbable to be due exclusively to sampling error. High heterogeneity is prevalent in educational meta-analyses, especially in studies concerning intricate phenomena like creativity (Means et al., \u003cspan citationid=\"CR37\" class=\"CitationRef\"\u003e2013\u003c/span\u003e), and its existence requires careful interpretation.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eBeyond quantifying aggregate effects, this study adopts a critical meta-analytic lens by examining the interplay of instructional and contextual moderators. This approach reflects an important shift in the meta-analytic tradition\u0026mdash;from asking \u003cem\u003ewhether\u003c/em\u003e an intervention works to interrogating \u003cem\u003efor whom\u003c/em\u003e, \u003cem\u003eunder what conditions\u003c/em\u003e, and \u003cem\u003ethrough what mechanisms\u003c/em\u003e it operates (Slavin, \u003cspan citationid=\"CR52\" class=\"CitationRef\"\u003e2020\u003c/span\u003e). Such analytical depth is essential in educational research, where interventions like PBL are inherently sensitive to implementation fidelity, curriculum alignment, and teacher competence (Harris \u0026amp; Hofer, \u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e2011\u003c/span\u003e).\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec19\" class=\"Section2\"\u003e \u003ch2\u003eThe Overall Effect of Project-Based Learning on Creative Thinking\u003c/h2\u003e \u003cp\u003eThe meta-analytic synthesis yielded a considerable and statistically significant effect size (SMD\u0026thinsp;=\u0026thinsp;0.8195), demonstrating that PBL consistently enhances students' creative thinking across diverse teaching situations. According to Cohen\u0026rsquo;s (\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e1988\u003c/span\u003e) criterion, this effect size signifies a pedagogically meaningful impact that extends beyond mere statistical significance. In the contemporary educational landscape, where creativity is recognized as a vital 21st-century skill, these findings hold both theoretical significance and practical necessity (Trilling \u0026amp; Fadel, \u003cspan citationid=\"CR60\" class=\"CitationRef\"\u003e2009\u003c/span\u003e). The findings affirm that PBL is not simply an instructional technique; it functions as a generative learning framework that cultivates diverse and evaluative thinking crucial for creativity.\u003c/p\u003e \u003cp\u003eThe effectiveness of PBL in fostering creative thinking aligns with socio-constructivist perspectives, particularly Vygotsky\u0026rsquo;s (1978) notion of the zone of proximal development and the need for mediated, socially contextualized learning. PBL emphasizes collaborative inquiry, real-world problem-solving, and iterative design, aligning with the cognitive demands of creative processes as outlined in Sternberg and Lubart's (1995) investment theory and Sawyer's (2014) model of disciplined improvisation. Unlike traditional teaching techniques that generally prioritize convergent thinking and rote memorization, PBL engages students in open-ended situations that require the generation, elaboration, and refinement of ideas\u0026mdash;critical components of creative cognition (Runco \u0026amp; Acar, \u003cspan citationid=\"CR46\" class=\"CitationRef\"\u003e2012\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eThe uniformity of beneficial outcomes across research supports the scalability of project-based learning (PBL) within educational settings. Although some educational innovations are confined to pilot initiatives or prestigious institutions, the extensive evidence in this synthesis indicates that project-based learning (PBL) can serve as a comprehensive innovation if executed with sufficient support and contextual awareness (Bell, \u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e2010\u003c/span\u003e). The creative advantages linked to PBL are not limited to high-achieving or affluent student groups; instead, the benefits are evident across all educational levels, from elementary to higher education.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec20\" class=\"Section2\"\u003e \u003ch2\u003eModerator Analysis\u003c/h2\u003e \u003cp\u003eThe moderating role of assessment instruments reveals a critical methodological insight: the way creative thinking is measured can substantially influence the estimated impact of PBL interventions. Structured assessments\u0026mdash;such as essays and standardized questionnaires\u0026mdash;demonstrate higher sensitivity in capturing the cognitive transformations fostered by PBL, particularly in areas like fluency, originality, and elaboration. These instruments benefit from predefined scoring rubrics and psychometric validation, which enhance their ability to detect meaningful changes across individuals and settings (Torrance, \u003cspan citationid=\"CR59\" class=\"CitationRef\"\u003e1974\u003c/span\u003e; Long et al., \u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e2022\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eIn contrast, less structured formats, such as open-ended questions, may introduce variability and scoring ambiguity that compromise measurement precision. Without standardized rubrics, such tools are prone to rater subjectivity and may underrepresent the nuanced effects of PBL on creative expression (Jonsson \u0026amp; Svingby, \u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e2007\u003c/span\u003e). Observation-based instruments, while offering valuable contextual insight into student behavior, pose additional risks of observer bias and often lack inter-rater reliability unless rigorously calibrated (Said-Metwaly et al., \u003cspan citationid=\"CR48\" class=\"CitationRef\"\u003e2017\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eThis finding underscores an urgent need for methodological convergence in assessing creative thinking, especially within complex learning models like PBL. As Condliffe (2017) notes, the absence of standardized and validated assessment tools in PBL research undermines the comparability and interpretability of findings. To advance the field, future studies must adopt or develop creativity assessments that are both theoretically grounded and empirically robust\u0026mdash;capable of capturing the multidimensional nature of creativity without sacrificing reliability or validity.\u003c/p\u003e \u003cp\u003eBeyond measurement considerations, the developmental stage of learners also plays a crucial role in shaping the effectiveness of PBL interventions. The moderating effect of educational level indicates that students\u0026rsquo; developmental stage significantly influences the impact of PBL on creative thinking. High school students tend to derive greater creative benefits from PBL interventions than their younger peers. This pattern is consistent with developmental research, which asserts that adolescence marks a shift toward more advanced forms of reasoning, including hypothetico-deductive logic, metacognition, and abstract problem-solving (Kuhn, \u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e2000\u003c/span\u003e). These higher-order cognitive abilities enable older students to navigate the open-ended, self-directed nature of PBL more effectively, allowing them to engage deeply with project goals, iterate on ideas, and generate original solutions.\u003c/p\u003e \u003cp\u003eElementary and middle school students may encounter challenges in sustaining inquiry over time or synthesizing interdisciplinary content without extensive scaffolding. According to Vygotsky (1978), younger learners function within a more limited zone of proximal development and thus require guided interaction and structured support to achieve complex cognitive tasks. Belland, Glazewski, and Ertmer (\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e2013\u003c/span\u003e) similarly emphasize that scaffolding is essential for promoting higher-level thinking among middle school students in PBL environments. Without adequate support, younger learners may struggle to articulate or externalize creative ideas, which can suppress observable gains in creativity.\u003c/p\u003e \u003cp\u003eThese findings underscore the importance of aligning PBL design with learners\u0026rsquo; developmental readiness. For younger students, project tasks may need to be more structured, with clearly defined phases and explicit prompts to guide creative thinking. Meanwhile, for older students, PBL can catalyze autonomous exploration and innovation. As Zimmerman (\u003cspan citationid=\"CR66\" class=\"CitationRef\"\u003e2002\u003c/span\u003e) notes, high school learners are typically more proficient in self-regulated learning\u0026mdash;a key enabler of sustained creative engagement within PBL settings. Future research should investigate how differentiated scaffolding strategies can be optimized across age groups to unlock the full creative potential of project-based instruction.\u003c/p\u003e \u003cp\u003eIn addition to learner age, the way PBL is delivered\u0026mdash;whether face-to-face, hybrid, or fully online\u0026mdash;emerges as another key factor influencing its creative impact. The type of PBL implementation significantly influenced its effectiveness in enhancing students\u0026rsquo; creative thinking, revealing that modality is not a neutral variable but a determinant of pedagogical impact. Traditional face-to-face PBL models consistently yielded the most robust outcomes, likely due to their ability to support embodied, socially situated learning\u0026mdash;hallmarks of creativity development. In such settings, students engage in immediate feedback cycles, physical prototyping, and spontaneous collaboration, conditions that foster iterative ideation and divergent thinking (Hmelo-Silver et al., \u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e2007\u003c/span\u003e; Thomas, \u003cspan citationid=\"CR57\" class=\"CitationRef\"\u003e2000\u003c/span\u003e). These findings reaffirm that creativity thrives in environments rich in interaction, contextual cues, and real-time social negotiation.\u003c/p\u003e \u003cp\u003ePjBL variants that emphasized concrete outputs maintained strong performance, possibly because they preserved the generative tension between product creation and process-oriented inquiry. However, the effectiveness of PBL dropped markedly in blended and especially fully online environments. The reduced effect sizes in digital formats suggest that key creative mechanisms may be attenuated or absent (Barak \u0026amp; Raz, \u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e2000\u003c/span\u003e; Means et al., \u003cspan citationid=\"CR36\" class=\"CitationRef\"\u003e2014\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eThese patterns highlight a pressing need for intentional design adaptation when implementing PBL in digital or hybrid contexts. Simply replicating offline project workflows online overlooks the epistemic shift required for digital creativity\u0026mdash;one that accounts for tool-mediated collaboration, multimodal expression, and scaffolded autonomy.\u003c/p\u003e \u003cp\u003eThe subject domain in which PBL is applied introduces unique affordances and constraints for fostering creativity. The subject domain in which PBL is implemented emerged as a significant moderator of its effectiveness in enhancing creative thinking, suggesting that disciplinary epistemologies and pedagogical affordances shape the extent to which creativity can be meaningfully expressed. PBL interventions in natural sciences, such as chemistry and biology, tended to produce larger effects compared to those in physics or integrated science. This variation may reflect differences in how each discipline conceptualizes inquiry, uncertainty, and problem space openness\u0026mdash;factors that influence the compatibility between domain content and creative engagement (Kind \u0026amp; Kind, \u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e2007\u003c/span\u003e; Barrow, \u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2006\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eIn subjects like chemistry and environmental science, where real-world phenomena are often complex, multifactorial, and open to multiple interpretations or solutions, students are more likely to encounter ill-structured problems that invite creative thinking. These contexts provide fertile ground for idea generation, design iteration, and the integration of interdisciplinary knowledge\u0026mdash;key conditions for creativity to flourish within PBL (Fortus et al., \u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e2005\u003c/span\u003e). By contrast, in disciplines such as physics, where problems are often more constrained, algorithmic, or deductive in nature, opportunities for creativity may be less explicit unless deliberately embedded through open-ended design tasks or contextualization strategies (Newton \u0026amp; Newton, 2011).\u003c/p\u003e \u003cp\u003eThese findings underscore the necessity of disciplinary adaptation in PBL design. While the model is inherently flexible, its creative affordances are not uniform across subjects. Therefore, Educators must design domain-sensitive PBL tasks that align with the nature of disciplinary knowledge and inquiry practices. For creativity to be meaningfully fostered, project prompts must move beyond content reproduction and invite students to question assumptions, explore alternatives, and synthesize across conceptual boundaries.\u003c/p\u003e \u003cp\u003eGeographic location significantly moderated the effectiveness of PBL interventions on students\u0026rsquo; creative thinking, pointing to the influence of broader sociocultural, educational, and systemic factors on how PBL is implemented and received. Studies conducted in Asian contexts tended to report larger effect sizes than those in Western or Middle Eastern regions. This pattern may be partially attributed to the increasing emphasis in many Asian education systems on pedagogical innovation and 21st-century skills\u0026mdash;often as part of national education reform agendas that position creativity as a strategic competence (OECD, \u003cspan citationid=\"CR42\" class=\"CitationRef\"\u003e2019\u003c/span\u003e; Cheng, 2010).\u003c/p\u003e \u003cp\u003eMoreover, the novelty of PBL in teacher-centered systems may have amplified its impact, as students in these settings are often unaccustomed to learning environments that promote autonomy, inquiry, and collaborative problem-solving. The contrast with traditional instruction could heighten engagement and perceived relevance, thereby stimulating creative responses. In contrast, in Western educational systems where constructivist or student-centered approaches are more prevalent, the marginal gain of PBL over existing practices may be less pronounced (Tan \u0026amp; Wang, \u003cspan citationid=\"CR56\" class=\"CitationRef\"\u003e2017\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eThese findings underscore that PBL is not a one-size-fits-all model; contextual readiness, cultural expectations, and systemic support mediate its effectiveness. Implementing PBL in diverse regions requires more than curricular insertion\u0026mdash;it demands adaptive strategies that consider local norms around creativity, authority, and learner autonomy. Future comparative studies should examine how policy environments, teacher beliefs, and institutional cultures interact to shape the outcomes of PBL interventions globally.\u003c/p\u003e \u003cp\u003eFinally, methodological characteristics such as sample size must be considered when interpreting the magnitude of observed effects. Sample size emerged as a significant moderator of the estimated impact of PBL on students\u0026rsquo; creative thinking, with studies employing smaller samples tending to report larger effect sizes. This trend aligns with well-documented patterns in educational and psychological research, wherein small-sample studies often yield inflated estimates due to greater sampling variability, increased susceptibility to bias, and reduced statistical power to detect heterogeneity (Button et al., \u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e2013\u003c/span\u003e; Sch\u0026auml;fer \u0026amp; Schwarz, \u003cspan citationid=\"CR50\" class=\"CitationRef\"\u003e2019\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eSmaller-scale studies may also involve more tightly controlled implementation conditions, such as direct researcher involvement, greater fidelity to PBL design, and more intensive scaffolding\u0026mdash;factors that may not generalize to larger classroom contexts. In contrast, larger-sample studies typically capture a broader range of learner diversity, instructional variation, and contextual complexity, which can dilute observed effects but enhance ecological validity (Slavin, \u003cspan citationid=\"CR51\" class=\"CitationRef\"\u003e2008\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eThese findings caution against overinterpreting high effect sizes from small samples without considering methodological rigor and external validity. Future PBL research should strive for larger, multisite designs with transparent reporting, while meta-analyses must continue to adjust for potential small-sample bias. Incorporating techniques such as precision weighting and sensitivity analyses can further mitigate the influence of sample size disparities and support more trustworthy generalizations.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec21\" class=\"Section2\"\u003e \u003ch2\u003ePractical Implications\u003c/h2\u003e \u003cp\u003eA central implication concerns the instrumental role of assessment fidelity. The differential effect sizes associated with structured versus unstructured assessment tools suggest that the observed success of PBL is not solely instructional but also a function of measurement precision. Instruments aligned with creativity constructs are essential for detecting the nuanced cognitive gains PBL intends to cultivate (Torrance, \u003cspan citationid=\"CR59\" class=\"CitationRef\"\u003e1974\u003c/span\u003e; Long et al., \u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e2022\u003c/span\u003e). Without such tools, evidence of impact may be either underreported or misinterpreted, undermining both instructional feedback and policy decisions.\u003c/p\u003e \u003cp\u003eEqually critical is the developmental alignment of PBL with learners' cognitive trajectories. The more substantial effects observed at the secondary level underscore that adolescents possess the metacognitive and self-regulatory capacities necessary to benefit from PBL\u0026rsquo;s open-ended and inquiry-driven nature (Kuhn, \u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e2000\u003c/span\u003e; Zimmerman, \u003cspan citationid=\"CR66\" class=\"CitationRef\"\u003e2002\u003c/span\u003e). For younger learners, whose executive functioning is still emerging, successful PBL requires deliberate scaffolding strategies that structure tasks without constraining creativity (Belland et al., \u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e2013\u003c/span\u003e; Vygotsky, 1978).\u003c/p\u003e \u003cp\u003eThe implementation modality of PBL also emerged as a consequential determinant of its creative impact. Face-to-face formats consistently facilitated richer, more generative learning environments than their digital counterparts. The diminished effectiveness observed in blended and online PBL signals a design gap: when stripped of physical collaboration, immediate feedback, and embodied interaction, PBL risks becoming procedural rather than epistemic (Barak \u0026amp; Raz, \u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e2000\u003c/span\u003e; Means et al., \u003cspan citationid=\"CR36\" class=\"CitationRef\"\u003e2014\u003c/span\u003e). Thus, digital adaptations must move beyond content replication and instead re-engineer the creative affordances of the physical classroom into virtual forms\u0026mdash;leveraging synchronous collaboration tools, design-based platforms, and feedback systems that preserve dialogic learning.\u003c/p\u003e \u003cp\u003eThe geographic and cultural context further mediates PBL\u0026rsquo;s success. In systems transitioning from didactic to student-centered models, PBL may yield amplified effects due to its novelty and perceived relevance (Cheng, 2010; Tan \u0026amp; Wang, \u003cspan citationid=\"CR56\" class=\"CitationRef\"\u003e2017\u003c/span\u003e). However, without adequate teacher preparation, curricular alignment, and institutional support, such gains may be short-lived. Policymakers and educational leaders must treat PBL not as a modular intervention but as a systemic shift\u0026mdash;requiring ecosystem-level reforms in instructional culture, assessment policy, and professional development (OECD, \u003cspan citationid=\"CR42\" class=\"CitationRef\"\u003e2019\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eFinally, the findings raise methodological imperatives for both researchers and evaluators. The inverse relationship between sample size and reported effect size highlights the danger of over-relying on small-N studies, which may reflect context-specific optimization rather than scalable effectiveness (Button et al., \u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e2013\u003c/span\u003e; Slavin, \u003cspan citationid=\"CR51\" class=\"CitationRef\"\u003e2008\u003c/span\u003e). Moving forward, PBL research must prioritize methodological transparency, replication, and multi-site trials to build a cumulative and generalizable evidence base.\u003c/p\u003e \u003c/div\u003e"},{"header":"Conclusion","content":"\u003cp\u003eThis meta-analysis confirms that PBL effectively enhances students\u0026rsquo; creative thinking in science education. Across 22 empirical studies, PBL demonstrated a substantial and statistically significant positive impact compared to traditional teaching methods. However, significant heterogeneity among studies suggests that its effectiveness depends on various contextual factors such as assessment type, learner level, instructional mode, subject area, location, and sample size.\u003c/p\u003e \u003cp\u003ePBL should be designed and implemented responsively to learners\u0026rsquo; developmental needs and local contexts to maximize creative outcomes. Future research should strengthen methodological rigor and explore long-term and cross-cultural applications. When thoughtfully implemented, PBL remains a powerful pedagogical framework for nurturing creativity as a core competency in 21st-century science education.\u003c/p\u003e \u003cdiv id=\"Sec23\" class=\"Section2\"\u003e \u003ch2\u003eLimitations and Directions for Future Research\u003c/h2\u003e \u003cp\u003eAlthough this meta-analysis demonstrates the positive impact of PBL on students\u0026rsquo; creative thinking in science education, several limitations must be noted. Creativity was measured using diverse and often unvalidated instruments, reducing comparability across studies. The data also show geographic and publication bias, with most studies concentrated in Asian contexts, limiting generalizability. Additionally, many interventions were short-term and involved small samples, making it difficult to assess the long-term sustainability of creativity gains.\u003c/p\u003e \u003cp\u003eFuture research should standardize creativity assessments, use larger and more diverse samples, and include longitudinal and cross-cultural designs. Developing validated, domain-specific tools and employing mixed-method approaches will help clarify how PBL fosters creativity over time and across contexts.\u003c/p\u003e \u003c/div\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eAcknowledgements\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eNot applicable.\u003c/p\u003e\n\n\u003cp\u003e\u003cstrong\u003eClinical trial number\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eNot applicable. This study is a meta-analysis and does not involve human participants, clinical interventions, or patient data collection; therefore, it is exempt from clinical trial registration requirements.\u003c/p\u003e\n\n\u003cp\u003e\u003cstrong\u003eEthics approval\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThis study did not involve human participants, animals, or biological material. Therefore, ethical approval was not needed.\u003c/p\u003e\n\n\u003cp\u003e\u003cstrong\u003eConsent to participate\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eNot applicable. This study did not include humans who consented to participate.\u003c/p\u003e\n\n\u003cp\u003e\u003cstrong\u003eConsent to publish\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eNot applicable. No individual person\u0026rsquo;s data in any form (including individual details, images, or videos) are included in the manuscript.\u003c/p\u003e\n\n\u003cp\u003e\u003cstrong\u003eAuthor contributions\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eIS led the conceptualization, designed the meta-analysis protocol, conducted the literature search and study screening, performed data extraction and statistical analysis, and drafted the initial manuscript. FR validated the search strategy, cross-checked study eligibility and data extraction, supported the analytical procedures, contributed to the interpretation of findings, and revised the manuscript critically for important intellectual content. BAP supervised the overall methodological framework, guided the interpretation and synthesis of results, provided substantial editorial input throughout the writing process, and approved the final version of the manuscript for submission.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eFunding Statement\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThis research received financial support from the Indonesian Education Scholarship (Beasiswa Pendidikan Indonesia), Center for Education Financial Services (Badan Layanan Umum \u0026ndash; LPDP), Ministry of Finance of the Republic of Indonesia, under Grant No. 00894/BPPT/BPI.06/9/2023. No additional external funding was received for this work.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\u003cli\u003e\u003cspan\u003eBarak, M., \u0026amp; Raz, E. (2000). Hot air balloons: Integrating web-based learning in elementary science. \u003cem\u003eJournal of Science Education and Technology\u003c/em\u003e, \u003cem\u003e9\u003c/em\u003e(1), 59\u0026ndash;67.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBarrow, L. H. (2006). A brief history of inquiry: From Dewey to standards. \u003cem\u003eJournal of Science Teacher Education\u003c/em\u003e, \u003cem\u003e17\u003c/em\u003e(3), 265\u0026ndash;278.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBeghetto, R. A., \u0026amp; Kaufman, J. C. (2017). \u003cem\u003eNurturing creativity in the classroom\u003c/em\u003e. Cambridge University Press.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBell, S. (2010). Project-based learning for the 21st century: Skills for the future. \u003cem\u003eThe Clearing House\u003c/em\u003e, \u003cem\u003e83\u003c/em\u003e(2), 39\u0026ndash;43.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBelland, B. R., Glazewski, K. D., \u0026amp; Ertmer, P. A. (2013). Problem-based learning and argumentation: Testing a scaffolding framework to support middle school students\u0026rsquo; creation of evidence-based arguments. \u003cem\u003eInstructional Science\u003c/em\u003e, \u003cem\u003e41\u003c/em\u003e(3), 473\u0026ndash;500.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBorenstein, M., Hedges, L. V., Higgins, J. P. T., \u0026amp; Rothstein, H. R. (2010). A basic introduction to fixed-effect and random‐effects models for meta-analysis. \u003cem\u003eResearch Synthesis Methods\u003c/em\u003e, \u003cem\u003e1\u003c/em\u003e(2), 97\u0026ndash;111.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eButton, K. S., Ioannidis, J. P., Mokrysz, C., Nosek, B. A., Flint, J., Robinson, E. S., \u0026amp; Munaf\u0026ograve;, M. R. (2013). Power failure: Why small sample size undermines the reliability of neuroscience. \u003cem\u003eNature Reviews Neuroscience\u003c/em\u003e, \u003cem\u003e14\u003c/em\u003e(5), 365\u0026ndash;376.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003e\u0026Ccedil;alik, M., \u0026amp; Wiyarsi, A. (2024). Exploring the effectiveness of project-based learning in promoting scientific creativity: A cross-cultural perspective. \u003cem\u003eJournal of Science Learning\u003c/em\u003e, \u003cem\u003e7\u003c/em\u003e(1), 11\u0026ndash;25.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eChen, C. H., \u0026amp; Yang, Y. C. (2019). Revisiting the effects of project-based learning on students\u0026rsquo; academic achievement: A meta-analysis investigating moderators. \u003cem\u003eEducational Research Review\u003c/em\u003e, \u003cem\u003e26\u003c/em\u003e, 71\u0026ndash;81. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.edurev.2018.11.001\u003c/span\u003e\u003cspan address=\"10.1016/j.edurev.2018.11.001\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCohen, J. (1988). \u003cem\u003eStatistical power analysis for the behavioral sciences\u003c/em\u003e (2nd ed.). Lawrence Erlbaum Associates.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCromley, J. G., Perez, T., \u0026amp; Kaplan, A. (2023). A meta-analysis of cognitive and motivational interventions in STEM education. \u003cem\u003eEducational Psychology Review\u003c/em\u003e, \u003cem\u003e35\u003c/em\u003e(1), 203\u0026ndash;225. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1007/s10648-022-09680-9\u003c/span\u003e\u003cspan address=\"10.1007/s10648-022-09680-9\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDemena, B. A. (2024). Detecting publication bias in educational meta-analyses: Challenges and recommendations. \u003cem\u003eInternational Journal of Educational Methodology\u003c/em\u003e, \u003cem\u003e10\u003c/em\u003e(1), 55\u0026ndash;71.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eEgger, M., Smith, G. D., Schneider, M., \u0026amp; Minder, C. (1997). Bias in meta-analysis detected by a simple, graphical test. \u003cem\u003eBmj\u003c/em\u003e, \u003cem\u003e315\u003c/em\u003e(7109), 629\u0026ndash;634. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1136/bmj.315.7109.629\u003c/span\u003e\u003cspan address=\"10.1136/bmj.315.7109.629\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eEgger, M., Smith, G. D., Schneider, M., \u0026amp; Minder, C. (1997). Bias in meta-analysis detected by a simple, graphical test. \u003cem\u003eBmj\u003c/em\u003e, \u003cem\u003e315\u003c/em\u003e(7109), 629\u0026ndash;634. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1136/bmj.315.7109.629\u003c/span\u003e\u003cspan address=\"10.1136/bmj.315.7109.629\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eFortus, D., Dershimer, R. C., Krajcik, J., Marx, R. W., \u0026amp; Mamlok-Naaman, R. (2005). Design-based science and real-world problem-solving. \u003cem\u003eInternational Journal of Science Education\u003c/em\u003e, \u003cem\u003e27\u003c/em\u003e(7), 855\u0026ndash;879.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGeng, J., \u0026amp; Su, Y. (2024). A meta-analysis of problem-based and project-based learning on STEM creativity in secondary education. \u003cem\u003eJournal of Educational Psychology\u003c/em\u003e, \u003cem\u003e116\u003c/em\u003e(1), 88\u0026ndash;104.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGillette, M. T., Gaffney, H., A. L., \u0026amp; Martin, J. M. (2018). Teaching for creativity: Models and practices in STEM classrooms. \u003cem\u003eInternational Journal of STEM Education\u003c/em\u003e, \u003cem\u003e5\u003c/em\u003e(1), 1\u0026ndash;15. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1186/s40594-018-0139-0\u003c/span\u003e\u003cspan address=\"10.1186/s40594-018-0139-0\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHarrer, M., Cuijpers, P., Furukawa, T. A., \u0026amp; Ebert, D. D. (2021). \u003cem\u003eDoing meta-analysis with R: A hands-on guide\u003c/em\u003e. Chapman \u0026amp; Hall/CRC. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://bookdown.org/MathiasHarrer/Doing_Meta_Analysis_in_R/\u003c/span\u003e\u003cspan address=\"https://bookdown.org/MathiasHarrer/Doing_Meta_Analysis_in_R/\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHarris, J., \u0026amp; Hofer, M. (2011). Technological pedagogical content knowledge (TPACK) in action. \u003cem\u003eJournal of Research on Technology in Education\u003c/em\u003e, \u003cem\u003e43\u003c/em\u003e(3), 211\u0026ndash;229.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHedges, L. V., Olkin, I., \u0026amp; Staton, T. (1989). Statistical methods for meta-analysis. \u003cem\u003eAcademic Press\u003c/em\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHmelo-Silver, C. E., Duncan, R. G., \u0026amp; Chinn, C. A. (2007). Scaffolding and achievement in problem-based and inquiry learning: A response to Kirschner, Sweller, and Clark (2006). \u003cem\u003eEducational Psychologist, 42\u003c/em\u003e(2), 99\u0026ndash;107. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1080/00461520701263368\u003c/span\u003e\u003cspan address=\"10.1080/00461520701263368\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHmelo-Silver, C. E., Duncan, R. G., \u0026amp; Chinn, C. A. (2007). Scaffolding and achievement in problem-based and inquiry learning: A response to Kirschner, Sweller, and Clark (2006). \u003cem\u003eEducational Psychologist, 42\u003c/em\u003e(2), 99\u0026ndash;107. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1080/00461520701263368\u003c/span\u003e\u003cspan address=\"10.1080/00461520701263368\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eJonsson, A., \u0026amp; Svingby, G. (2007). The use of scoring rubrics: reliability, validity, and educational consequences. \u003cem\u003eEducational Research Review\u003c/em\u003e, \u003cem\u003e2\u003c/em\u003e(2), 130\u0026ndash;144.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKim, H., Park, Y., \u0026amp; Baek, Y. (2020). Not just fun, but serious strategies: Using meta-analysis to explore effective game-based learning in STEM education. \u003cem\u003eEducational Technology Research and Development\u003c/em\u003e, \u003cem\u003e68\u003c/em\u003e(5), 2301\u0026ndash;2325. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1007/s11423-020-09796-z\u003c/span\u003e\u003cspan address=\"10.1007/s11423-020-09796-z\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKind, P., \u0026amp; Kind, V. (2007). Creativity in science education: Perspectives and challenges. \u003cem\u003eThinking Skills and Creativity\u003c/em\u003e, \u003cem\u003e2\u003c/em\u003e(1), 22\u0026ndash;29. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.tsc.2006.10.001\u003c/span\u003e\u003cspan address=\"10.1016/j.tsc.2006.10.001\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKind, P., \u0026amp; Kind, V. (2007). Creativity in science education: Perspectives and challenges. \u003cem\u003eThinking Skills and Creativity\u003c/em\u003e, \u003cem\u003e2\u003c/em\u003e(1), 22\u0026ndash;29.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKokotsaki, D., Menzies, V., \u0026amp; Wiggins, A. (2016). Project-based learning: A review of the literature. \u003cem\u003eImproving Schools\u003c/em\u003e, \u003cem\u003e19\u003c/em\u003e(3), 267\u0026ndash;277. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1177/1365480216659733\u003c/span\u003e\u003cspan address=\"10.1177/1365480216659733\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKolaski, M., Jensen, T., \u0026amp; Rauf, A. (2023). Exploring meta-analytic trends in creative education research: A comprehensive synthesis. \u003cem\u003eCreativity Research Journal\u003c/em\u003e, \u003cem\u003e35\u003c/em\u003e(2), 123\u0026ndash;141.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKooren, E., Kicken, W., Loyens, S., \u0026amp; Merri\u0026euml;nboer, J. (2024). A meta-analysis of the effectiveness of design-based learning in secondary science education. \u003cem\u003eJournal of Research in Science Teaching\u003c/em\u003e, \u003cem\u003e61\u003c/em\u003e(1), 33\u0026ndash;59.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKuhn, D. (2000). Metacognitive development. \u003cem\u003eCurrent Directions in Psychological Science\u003c/em\u003e, \u003cem\u003e9\u003c/em\u003e(5), 178\u0026ndash;181.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLe Thi Tuyet, N., Hoang, T. M., \u0026amp; Vu, M. L. (2024). Meta-analysis of inquiry-based learning on students\u0026rsquo; critical thinking in science classrooms. \u003cem\u003eInternational Journal of Science Education\u003c/em\u003e, \u003cem\u003e46\u003c/em\u003e(2), 150\u0026ndash;170.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLi, Y., Wang, K., \u0026amp; Xiao, Y. (2020). The effects of project-based learning on students' academic performance: A meta-analysis. \u003cem\u003eEducational Research and Evaluation\u003c/em\u003e, \u003cem\u003e26\u003c/em\u003e(1\u0026ndash;2), 89\u0026ndash;105. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1080/13803611.2020.1784855\u003c/span\u003e\u003cspan address=\"10.1080/13803611.2020.1784855\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLinden, J., M\u0026uuml;ller, R., \u0026amp; Chang, K. (2024). Funnel plot asymmetry and publication bias in science education: A simulation-based analysis. \u003cem\u003eEducational Measurement \u0026amp; Evaluation Review\u003c/em\u003e, \u003cem\u003e29\u003c/em\u003e(1), 23\u0026ndash;41.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLong, H., Kerr, B., Emler, T., \u0026amp; Birdnow, M. (2022). A critical review of assessments of creativity in education. \u003cem\u003eReview of Research in Education\u003c/em\u003e, \u003cem\u003e46\u003c/em\u003e(1), 288\u0026ndash;323.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMalapane, M., White, P., \u0026amp; Tekkumru-Kisa, M. (2022). Supporting teachers to teach for creativity in science: A review of the literature. \u003cem\u003eStudies in Science Education\u003c/em\u003e, \u003cem\u003e58\u003c/em\u003e(3), 253\u0026ndash;280. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1080/03057267.2021.1952382\u003c/span\u003e\u003cspan address=\"10.1080/03057267.2021.1952382\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMeans, B., Toyama, Y., Murphy, R., \u0026amp; Bakia, M. (2014). Learning online: What research tells us about whether, when, and how. \u003cem\u003eRoutledge\u003c/em\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMeans, B., Toyama, Y., Murphy, R., Bakia, M., \u0026amp; Jones, K. (2013). \u003cem\u003eThe effectiveness of online and blended learning: A meta-analysis of the empirical literature\u003c/em\u003e. U.S. Department of Education.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eNakagawa, Y., Lee, S., \u0026amp; Tran, T. (2023). A meta-analysis of PBL effectiveness in East Asian science classrooms. \u003cem\u003eAsia-Pacific Education Researcher\u003c/em\u003e, \u003cem\u003e32\u003c/em\u003e(1), 45\u0026ndash;63.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eNakagawa, Y., Lee, S., \u0026amp; Tran, T. (2023). A meta-analysis of PBL effectiveness in East Asian science classrooms. \u003cem\u003eAsia-Pacific Education Researcher\u003c/em\u003e, \u003cem\u003e32\u003c/em\u003e(1), 45\u0026ndash;63.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eNakagawa, Y., Lee, S., \u0026amp; Tran, T. (2023). A meta-analysis of PBL effectiveness in East Asian science classrooms. \u003cem\u003eAsia-Pacific Education Researcher\u003c/em\u003e, \u003cem\u003e32\u003c/em\u003e(1), 45\u0026ndash;63.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eNeethirajan, S., Park, J., \u0026amp; Kelton, D. (2005). Publication bias and funnel plot asymmetry in veterinary epidemiology. \u003cem\u003ePreventive Veterinary Medicine\u003c/em\u003e, \u003cem\u003e69\u003c/em\u003e(2\u0026ndash;4), 165\u0026ndash;172.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eOECD. (2019). \u003cem\u003eOECD Learning Compass 2030: A series of concept notes\u003c/em\u003e. OECD Publishing. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://www.oecd.org/education/2030-project/\u003c/span\u003e\u003cspan address=\"https://www.oecd.org/education/2030-project/\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003e\u0026Ouml;zt\u0026uuml;rk, N., Bayram, H., \u0026amp; G\u0026uuml;ler, M. (2022). The effects of project-based learning on students\u0026rsquo; creative thinking: A meta-analysis. \u003cem\u003eJournal of Educational Research and Practice\u003c/em\u003e, \u003cem\u003e12\u003c/em\u003e(4), 401\u0026ndash;420.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRaposo-Rivas, M., Mart\u0026iacute;nez-Figueira, E., \u0026amp; Sarmiento Campos, J. A. (2024). The effect of project-based learning on students\u0026rsquo; higher-order thinking skills: A systematic review and meta-analysis. \u003cem\u003eEducational Research Review\u003c/em\u003e, \u003cem\u003e38\u003c/em\u003e, 100529.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRaposo-Rivas, M., Mart\u0026iacute;nez-Figueira, E., \u0026amp; Sarmiento Campos, J. A. (2024). The effect of project-based learning on students\u0026rsquo; higher-order thinking skills: A systematic review and meta-analysis. \u003cem\u003eEducational Research Review\u003c/em\u003e, \u003cem\u003e38\u003c/em\u003e, 100529.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRunco, M. A., \u0026amp; Acar, S. (2012). Divergent thinking as an indicator of creative potential. \u003cem\u003eCreativity Research Journal\u003c/em\u003e, \u003cem\u003e24\u003c/em\u003e(1), 66\u0026ndash;75.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRunco, M. A., \u0026amp; Jaeger, G. J. (2012). The standard definition of creativity. \u003cem\u003eCreativity Research Journal\u003c/em\u003e, \u003cem\u003e24\u003c/em\u003e(1), 92\u0026ndash;96. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1080/10400419.2012.650092\u003c/span\u003e\u003cspan address=\"10.1080/10400419.2012.650092\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSaid-Metwaly, S., Van den Noortgate, W., \u0026amp; Kyndt, E. (2017). Approaches to measuring creativity: A systematic literature review. \u003cem\u003eCreativity Theories\u0026mdash;Research\u0026mdash;Applications\u003c/em\u003e, \u003cem\u003e4\u003c/em\u003e(2), 238\u0026ndash;275.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSavery, J. R. (2015). Overview of problem-based learning: Definitions and distinctions. \u003cem\u003eInterdisciplinary Journal of Problem-Based Learning\u003c/em\u003e, \u003cem\u003e1\u003c/em\u003e(1), 9\u0026ndash;20. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.7771/1541-5015.1002\u003c/span\u003e\u003cspan address=\"10.7771/1541-5015.1002\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSch\u0026auml;fer, T., \u0026amp; Schwarz, M. A. (2019). The meaningfulness of effect sizes in psychological research: Differences between academic sub-disciplines and the impact of publication bias. \u003cem\u003eFrontiers in Psychology\u003c/em\u003e, \u003cem\u003e10\u003c/em\u003e, 813.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSlavin, R. E. (2008). Perspectives on evidence-based research in education\u0026mdash;What works? Issues in synthesizing educational program evaluations. \u003cem\u003eEducational Researcher\u003c/em\u003e, \u003cem\u003e37\u003c/em\u003e(1), 5\u0026ndash;14.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSlavin, R. E. (2020). How evidence-based reform will transform research and practice in education. \u003cem\u003eEducational Psychologist\u003c/em\u003e, \u003cem\u003e55\u003c/em\u003e(1), 21\u0026ndash;31.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSmela, K., Rodriguez, M., \u0026amp; Tatum, D. (2023). Measuring the effectiveness of project-based learning on scientific creativity: A meta-synthesis. \u003cem\u003eInternational Journal of Science Education\u003c/em\u003e, \u003cem\u003e45\u003c/em\u003e(4), 512\u0026ndash;534.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSterne, J. A. C., Egger, M., \u0026amp; Moher, D. (2011). Addressing reporting biases. In J. P. T. Higgins \u0026amp; S. Green (Eds.), \u003cem\u003eCochrane Handbook for Systematic Reviews of Interventions\u003c/em\u003e (Version 5.1.0). The Cochrane Collaboration.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSuprapto, N., Liu, Y., \u0026amp; Ku, C. H. (2021). The effectiveness of project-based learning on students' science academic achievement: A meta-analysis study. \u003cem\u003eJournal of Baltic Science Education\u003c/em\u003e, \u003cem\u003e20\u003c/em\u003e(1), 90\u0026ndash;104. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.33225/jbse/21.20.90\u003c/span\u003e\u003cspan address=\"10.33225/jbse/21.20.90\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eTan, O. S., \u0026amp; Wang, T. (2017). \u003cem\u003eProblem-based learning innovation: Using problems to power learning in the 21st century\u003c/em\u003e. Springer.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eThomas, J. W. (2000). \u003cem\u003eA review of research on project-based learning\u003c/em\u003e. The Autodesk Foundation.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eTing, C., Huang, L., \u0026amp; Zhao, W. (2023). Creativity-focused pedagogies in science education: A cross-national meta-analysis. \u003cem\u003eScience Education\u003c/em\u003e, \u003cem\u003e107\u003c/em\u003e(5), 1123\u0026ndash;1142.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eTorrance, E. P. (1974). \u003cem\u003eTorrance tests of creative thinking: Norms-technical manual\u003c/em\u003e. Personnel.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eTrilling, B., \u0026amp; Fadel, C. (2009). \u003cem\u003e21st-century skills: Learning for life in our times\u003c/em\u003e. Jossey-Bass.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWen, M. L., Tsai, C. C., \u0026amp; Chang, C. Y. (2015). Cognitive load and learning effect in STEM project-based learning: A meta-analysis. \u003cem\u003eJournal of Educational Technology \u0026amp; Society\u003c/em\u003e, \u003cem\u003e18\u003c/em\u003e(4), 34\u0026ndash;47.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eYang, Y., Chen, P., \u0026amp; Zhao, S. (2020). The impact of project-based learning on students\u0026rsquo; higher-order thinking: A meta-analysis. \u003cem\u003eThinking Skills and Creativity\u003c/em\u003e, \u003cem\u003e38\u003c/em\u003e, 100718.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eYu, C., \u0026amp; Xu, H. (2022). A meta-analysis of project-based learning effectiveness in fostering students\u0026rsquo; creative outcomes. \u003cem\u003eEducational Psychology Review\u003c/em\u003e, \u003cem\u003e34\u003c/em\u003e, 1015\u0026ndash;1034. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1007/s10648-021-09605-4\u003c/span\u003e\u003cspan address=\"10.1007/s10648-021-09605-4\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZhang, L., Tan, W., \u0026amp; Li, M. (2024). Reassessing creativity in STEM education: Meta-analytic insights into instructional interventions. \u003cem\u003eJournal of Creative Behavior\u003c/em\u003e, \u003cem\u003e58\u003c/em\u003e(2), 211\u0026ndash;229.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZhang, L., Tan, W., \u0026amp; Li, M. (2024). Reassessing creativity in STEM education: Meta-analytic insights into instructional interventions. \u003cem\u003eJournal of Creative Behavior\u003c/em\u003e, \u003cem\u003e58\u003c/em\u003e(2), 211\u0026ndash;229.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZimmerman, B. J. (2002). Becoming a self-regulated learner: An overview. \u003cem\u003eTheory Into Practice\u003c/em\u003e, \u003cem\u003e41\u003c/em\u003e(2), 64\u0026ndash;70.\u003c/span\u003e\u003c/li\u003e\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":true,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true},"keywords":"Creative thinking, educational strategies, project-based learning, science education, meta-analysis","lastPublishedDoi":"10.21203/rs.3.rs-8231968/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-8231968/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eFostering students\u0026rsquo; creative thinking is a central goal of contemporary science education, and Project-Based Learning (PBL) has been widely promoted as a promising pedagogical approach to achieve it. The overall magnitude and consistency of PBL\u0026rsquo;s impact on creative thinking have not been comprehensively quantified. This study bridges that gap by conducting a systematic meta-analysis of 22 empirical studies published between 2014 and 2024, examining the effects of PBL on students\u0026rsquo; creative thinking across diverse educational levels. The random-effects model revealed a significant positive effect of PBL compared with traditional instruction, with a pooled standardized mean difference (SMD) of 0.82 and a 95% confidence interval [0.42, 1.22]. The heterogeneity index (I\u0026sup2; = 92.1%) indicated considerable variability among study outcomes. Moderator analyses further showed that the effect size differed significantly according to assessment type, educational level, instructional model, subject domain, geographic region, and sample size, with the strongest effects observed among high school students. These findings provide robust empirical support for the integration of PBL into science curricula to foster creative thinking. The results highlight the need for educators and policymakers to adapt PBL implementation to local educational contexts to maximize its effectiveness and sustainability.\u003c/p\u003e","manuscriptTitle":"Reimagining Science Education through Project-Based Learning: A Systematic Review of Creative Thinking Outcomes","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2026-02-09 09:38:09","doi":"10.21203/rs.3.rs-8231968/v1","editorialEvents":[{"type":"communityComments","content":0}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"403352fc-e0dc-4122-b5cc-8de82ef716d4","owner":[],"postedDate":"February 9th, 2026","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[],"tags":[],"updatedAt":"2026-03-07T12:25:06+00:00","versionOfRecord":[],"versionCreatedAt":"2026-02-09 09:38:09","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-8231968","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-8231968","identity":"rs-8231968","version":["v1"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}