Preparing students for evolving science, technology, engineering, and mathematics (STEM) careers requires teachers who understand workforce skills and pathways, yet most professional development (PD) and STEM Experiences for Teachers (SETs) are studied in isolation. This study examines the effects of the Kenan Fellows Program for Teacher Leadership (KFP) on teachers’ professional growth. KFP is a year-long model that integrates a 120h summer industry or research immersion with 80h of pedagogical and leadership PD spread across summer, fall, and spring institutes. We conducted a secondary analysis of transcripts from 56 interviews and 3 focus-groups from 73 teachers who participated in KFP between 2018 and 2024. The data were originally collected for the purpose of program evaluation. We employed inductive coding and then mapped emergent categories to Clarke and Hollingsworth’s (2002) Interconnected Model of Professional Growth (IMPG) to examine professional growth in the participating teachers’ personal domain, domain of practice, and domain of consequence. The IMPG accounts for the complex and multi-faceted nature of teacher professional growth that flows from PD experiences, providing a useful lens for examining the effects of KFP on teacher growth. Our findings indicate that the participating teachers most frequently reported changes in their personal domain (n = 371), with more references to knowledge (n = 160) than beliefs (n = 128) or attitudes (n = 69). Growth in their domain of practice (n = 211) centered on pedagogy (n = 136) and forming new community connections (n = 54). Reported changes in their domain of consequence (n = 241) highlighted teacher outcomes (n = 162) and new student opportunities (n = 95). Co-occurrence analyses linked knowledge gains to professional development and industry connections (72 co-occurrences) and belief change to program infrastructure/values and brand recognition (52 co-occurrences). A salient pathway traced increased self-efficacy → agentic action → new community connections → opportunities for students. The results of this study highlight how integrating SETs with sustained pedagogical and leadership PD drives multi-domain teacher growth and bridges the “knowing–doing gap” often seen in professional learning. Findings show that KFP’s SET + PD model enhances teachers’ knowledge, self-efficacy, and networks, which in turn catalyze agentic actions that expand classroom opportunities for students (e.g., classroom projects connected to apprenticeship programs). Through the lens of Clarke and Hollingsworth’s IMPG, the study demonstrates that external supports—such as industry experiences, high-quality PD, and program infrastructure—produce personal growth that translates into new pedagogical practices, leadership roles, and student engagement. This study positions the KFP model as a replicable framework for workforce-aligned teacher professional development and offers design and policy recommendations for scaling SET + PD initiatives to strengthen K–12 STEM education and career readiness.
Understanding STEM identity as process oriented may clarify how students form commitments to STEM and how these processes shape motivation. This study examined three STEM identity development processes—commitment, in-depth exploration, and reconsideration—and tested how they relate to achievement motivation. Undergraduate students enrolled in engineering courses (N = 230) completed an adapted Utrecht-Management of Identity Commitments Scale (U-MICS) and achievement-value scales. Confirmatory factor analyses supported a 12-item, three-factor model; multiple indicators multiple causes modeling (MIMIC) analyses indicated that older students reported more exploration and less reconsideration, and women reported higher commitment. Structural models indicated that commitment and exploration were positively associated with intrinsic and combined attainment–utility value, whereas reconsideration was negatively associated with intrinsic and attainment–utility value and positively associated with cost. Findings support a process-oriented view of STEM identity and demonstrate that identity processes differentially correspond to students’ motivational beliefs as they navigate early STEM pathways.
Gender imbalance in STEM, characterized by a significant underrepresentation of women, remains a significant challenge. Although gender bias is a well-known contributor to women’s attrition from STEM, the psychological mechanisms linking gender bias to departure are less well understood. Our research investigates early antecedents of attrition and the psychological processes that precede leaving the STEM pathway in tertiary education. Drawing upon identity integration research, we propose that perceived gender bias exerts undermines female STEM students’ academic performance and career aspirations by reducing their Gender-Professional Identity Integration (G-PII), a construct that captures individual differences in the perceived compatibility between a woman’s gender identity and her professional identity within a male-dominated STEM field. Across two studies, we examined 344 college students from the field of Computing and Information Sciences (CIS). In a cross-sectional study (Study 1), we first demonstrated that gender moderates the relationship between perceived gender bias and G-PII among STEM students. We then conducted a longitudinal study (Study 2) to further examine this moderation effect, assessing how perceived gender bias in CIS influences G-PII, in turn, influences female students’ academic performance and future career aspirations in the field. This pathway is not observed among their male counterparts. Our findings showed that the perceived bias—G-PII—career aspirations/academic performance link emerged only among female students. This highlights how gender bias within STEM fields undermines the perceived congruity between female students’ gender and their STEM identities, negatively impacting their academic performance and reducing their interest in pursuing STEM careers after graduation. These insights reveal the psychological factors driving women’s attrition in STEM post-college and highlight the need for educators and practitioners to address gender bias through an identity-focused lens. Strengthening female students’ internal congruity through interventions that increase G-PII may enhance their academic performance and confidence in pursuing STEM careers.
The rapid proliferation of generative Artificial Intelligence (AI) into secondary education has outpaced formal institutional guidance, creating an expansive array of unsupervised, independent student digital practices. Moving beyond simplistic assumptions of technological affordances, this cross-sectional study empirically analyses the self-reported behavioral distributions of secondary students’ interaction with generative AI-driven tools across four STEM subjects: computer science, mathematics, natural sciences, and economics. Using a convergent parallel mixed-methods design, we synthesized survey data (n = 416) and thematic reflections from business-oriented secondary schools in the Czech Republic to examine how institutional regulation and subject-specific environments shape student adoption. Quantitative findings reveal a stratified, divergent adoption model characterizing distinct academic fields when treated as structural and environmental proxies for varying disciplinary epistemologies and task environments. While applied and computational disciplines demonstrate normalized integration, theoretically rigorous subjects like mathematics exhibit a transparency gap characterized by high perceived prohibition and persistent clandestine use. We map the self-reported functional roles of AI in the student workflow, finding that students predominantly position AI as an instrumental scaffold for explanation and procedural verification rather than a direct substitute for independent reasoning. However, qualitative evidence uncovers a critical evaluation gap that exposes a behavioral asymmetry within the student workflow: operational prompting loops and epistemic verification are shown to function not as independent psychometric traits, but as interdependent workflow phases where continuous prompt modification heavily overshadows external factual validation. Although students frequently encounter algorithmic errors and procedural dissonance, where AI logic diverges from curricular standards, systematic cross-verification remains reported at low frequencies. The study suggests that AI integration is currently evolutionary rather than disruptive, often functioning as a cognitive mediator that accelerates workflow while shifting the behavioral layout toward operational dependency. These associational patterns highlight the need to transition from restrictive governance, which faces distinct reverse-causality reporting biases, to a framework that formally integrates overlapping operational and epistemic workflow competencies, emphasizing the need for subject-sensitive guidance that ensures AI enhances rather than displaces disciplinary rigor.
Pre-service teachers’ acceptance of generative AI is highly heterogeneous, yet most studies rely on variable-centered approaches that cannot reveal qualitatively distinct subgroups. This study adopted a person-centered latent profile analysis (LPA) to identify pre-service teachers’ ChatGPT acceptance profiles and to compare STEM and non-STEM teachers in terms of profile distribution and behavioral intention. Data were collected from N = 128 pre-service teachers in Taiwan (68 STEM, 60 non-STEM). Building on the Technology Acceptance Model (TAM) and Unified Theory of Acceptance and Use of Technology 2 (UTAUT2), five constructs were used as LPA indicators, with behavioral intention treated as a distal outcome. The analysis identified four highly distinguishable profiles (entropy = 0.985): Pragmatic Evaluators (47.66
In the modern digital era, computational thinking (CT) is a fundamental skill that has been influenced by advances in artificial intelligence (AI). While there are various resources for teaching and evaluating CT, the integration of AI principles into CT education remains scarce. In this study, we proposed the Computational Thinking in AI Training Test (CTAT), tailored for higher education students in computer sciences (CS) and AI-related programs. The test was developed using the Evidence-Centered Design (ECD) framework, resulting in 34 multiple-choice items being created. Validation was conducted through expert review and cognitive interviews, with refinements made based on feedback. A field test was then conducted with 461 higher education students (Year 1–2) from two vocational schools and two colleges in southern China. The data were analyzed using the Item Response Theory (IRT) approach, along with other psychometric tests. Results indicate that CTAT exhibited robust psychometric properties, suggesting its appropriateness for the target student population. Analysis of test performance revealed that students’ CT skills improved as they progressed through their educational pathways, with no significant gender-based differences observed. Students’ response patterns to the items reveal that they tend to have difficulties in identifying appropriate data representation, applying logical operators in a correct sequence, and differentiating loop structures. The study provides a valid and reliable tool for assessing CT within an AI training context at the higher education level, and findings on students’ challenges in solving CT problems provide practical insights for CT and AI tertiary education.
Abstract As K-12 education prepare learners for an AI-driven society, the interdisciplinary STEAM approach is widely presented as a pathway to AI literacy, yet evidence on where, how, and how well it advances distinct literacy areas remains fragmented. Following PRISMA procedures, we systematically reviewed 39 studies (2016–2025) from four databases that addressed AI as content within STEAM contexts. We conducted (1) descriptive mapping of publication year, region, educational level, research design, and instructional modality; and (2) deductive coding with the TIECD framework (Technology, Impact, Ethics, Collaboration, Design), followed by inductive refinement into ten AI Literacy Elements (AILEs). Co-occurrence analyses examined emphases and gaps, and mappings linked AILEs to school subjects and to STEAM clusters. Results show a sharp increase in publications from 2021 onward, with most work in middle and high schools. The corpus is geographically concentrated; methods are mainly mixed or qualitative; and instruction is predominantly technology-enhanced. Across AILEs, technical foundations, specifically Fundamental AI Concepts, Computational Thinking, and Data Literacy dominate, while Ethical Awareness, Creative Imagination, Creating with AI, Managing AI, and Designing AI are comparatively underrepresented. Contributions are led by Technology disciplines (Computer Science and Data Science), with thinner coverage across Science, Mathematics, Engineering and Arts disciplines, and Integrated STEAM. Overall, current STEAM implementations chiefly develop technical literacies while offering limited, uneven opportunities for ethical reasoning, creative futures thinking, collaborative management of AI, and design of AI systems. We propose a revised, evidence-informed AI literacy framework aligning TIECD with ten elements and recommend broadening early-years provision, diversifying disciplinary pathways beyond Technology, and designing tasks that jointly evidence technical fluency, ethics, collaboration with AI, and iterative design.
Transfer students are a growing group within U.S. higher education, yet their post‑transfer trajectories in Science, Technology, Engineering, and Mathematics (STEM) fields are understudied. Using a longitudinal dataset of 3416 transfer students who earned a bachelor’s degree from a large public university in Louisiana (2012–2018), we examined patterns of STEM degree attainment among all transfer students, STEM joiners (those who entered a STEM major after transfer) and STEM leavers (those who left a STEM major for a non‑STEM degree). The study aimed to identify demographic, pre‑transfer, and first‑year academic predictors of these distinct pathways, thereby informing institutional policies to support diverse transfer populations. Descriptive and logistic regression analyses revealed unique patterns and significant predictors of STEM degree attainment for all transfer students, STEM joiners, and STEM leavers. STEM degree recipients tended to arrive with stronger academic foundations, higher early-year GPAs, and more intense course loads than non-STEM graduates. Individuals who transferred laterally between four-year institutions and those who already held a prior bachelor’s degree were particularly likely to finish in STEM. In particular, our examination of STEM joiners and STEM leavers highlights the complex and fluid nature of STEM trajectories and the potential role of transfer pathways in diversifying STEM. Transfer students exhibit heterogeneous STEM trajectories shaped by both pre-existing academic capital and early campus experiences. The findings highlight the role of transfer pathways, particularly from community colleges, in increasing diversity in STEM fields. Institutions and policymakers should focus on strengthening these pathways, especially for women and underrepresented minority students. This could involve targeted outreach programs, mentorship initiatives, and increased funding for transfer student support services in STEM fields. By doing so, institutions can enhance STEM persistence and completion among transfer populations, contributing to a more diverse and skilled future workforce.
Problem posing serves as the starting point for inquiry-based learning in science, technology, engineering, and mathematics (STEM) education. However, it has not received sufficient attention in primary science education practices. Although the widespread adoption of generative artificial intelligence (GenAI) presents new opportunities to improve this situation, empirical evidence regarding how it facilitates problem posing in inquiry-based learning remains limited. This study employed a quasi-experimental design to systematically compare the effects of GenAI-based chatbots and search engines on primary school students’ problem posing ability in science. A total of 97 third-grade students from a primary school in central China were randomly assigned by class to the experimental group (N = 48, using a GenAI-based chatbot) and the control group (N = 49, using a search engine). Following three rounds of progressive problem-based inquiry learning, students’ problem posing ability, cognitive network structures, and learning experiences were analyzed and evaluated. Results showed that the use of chatbots significantly improved problem quality (t = 2.47, p = 0.015) and overall problem posing ability (t = 3.07, p = 0.003). Epistemic network analysis further showed that the experimental group developed a more integrated network structure, whereas the control group exhibited a pattern of local clustering. In addition, the experimental group reported lower cognitive load and higher technology acceptance. Overall, GenAI-based chatbots significantly outperformed search engines in facilitating problem posing. This study reveals the potential of GenAI in STEM education, providing empirical evidence for introducing GenAI tools into inquiry-based learning in primary science classrooms.
This study examines the effectiveness of an adaptive artificial intelligence–based instructional program grounded in the STEM approach in developing deep learning (defined here as students’ deep learning rather than machine learning algorithms) in science among sixth-grade students within a female-only primary school context. Using a concurrent mixed-methods design, the study integrated quantitative and qualitative evidence to evaluate how adaptive AI-driven learning pathways influenced students’ explanation, interpretation, application, and idea-generation skills. A cluster-randomized posttest-only control group design was implemented with 30 students assigned at the intact classroom level (one classroom per condition); therefore, findings are best interpreted as preliminary, within-sample evidence consistent with a program effect, and statistical estimates are treated as exploratory at the student level. The adaptive AI-based STEM program—designed to personalize content, adjust task difficulty, and provide real-time feedback—was implemented over eight weeks, with adaptive learning pathways governed by rule-based mastery mechanisms and machine-learning components used solely for performance monitoring rather than direct modification of learning trajectories. Quantitative results revealed statistically significant differences, based on median (IQR) summaries and Mann–Whitney U tests, favoring the experimental group across all deep learning dimensions, including explanation, interpretation, application, and idea generation. Large within-sample effect sizes indicated substantial rank separation within this sample, although classroom-level confounding and baseline differences cannot be fully ruled out in a two-cluster design. Qualitative findings from interviews with three science teachers corroborated these results, highlighting perceived improvements in analytical reasoning, conceptual integration, inquiry-based exploration, and creative scientific thinking. Teachers also reported that simulations, adaptive feedback, and hands-on STEM activities supported students’ conceptual understanding and engagement with scientific problem solving. Collectively, the findings provide context-specific pilot evidence suggesting that adaptive AI-based STEM instructional models may support deep learning in elementary science. The integration of adaptive AI mechanisms with inquiry-oriented STEM learning appears to promote deeper conceptual explanation, scientific interpretation, real-world application, and idea generation. However, because the study was conducted within a single female-only school using two intact classrooms, the findings should be interpreted cautiously and validated through larger, multi-site studies with stronger cluster-level experimental controls.
Despite substantial evidence supporting the efficacy of evidence-based study strategies in enhancing student learning outcomes, research reporting their integration into Science, Technology, Engineering, and Mathematics (STEM) higher education courses remains limited. This gap often arises from educators’ limited familiarity with study strategy practices, as well as the challenges in adapting general principles to diverse disciplinary contexts. To address this challenge, this systematic review examines the integration of six evidence-based study strategies – retrieval practice, spacing, interleaving, elaboration, concrete examples, and dual coding – within STEM higher education. Using the PRISMA protocol, 65 articles, selected from 2515 studies, were analyzed. Reviewed literature has demonstrated that undergraduate students who engage regularly and holistically with evidence-based study strategies demonstrated significantly improved learning outcomes and overall academic success. Our findings revealed a predominant focus on assessment-driven implementations, highlighting educators’ reliance on students’ extrinsic motivation, real or perceived. Challenges such as student resistance to new study strategies, logistical constraints, and variability in student readiness were prevalent. Notable gaps in the research include methodological limitations and a lack of longitudinal data. This review synthesizes literature that illustrates how educators have successfully contextualized evidence-based study strategies into their STEM undergraduate courses, offering a practical resource to facilitate broader adoption and adaptation by others in the field. It underscores the need for deliberate integration of study strategies, enhancement of student metacognition, and adaptation to diverse learner needs. Future research into integration of evidence-based study strategies into STEM higher education should employ robust experimental designs and explore the sustainability of interventions over time. This work contributes to the discourse on effective pedagogical practices in STEM higher education by providing insights into the implementation and impact of evidence-based study strategies.
To address the growing interest in, yet limited understanding of, ChatGPT-supported inquiry-based learning in STEAM education, this study systematically reviewed 24 empirical articles to identify current research trends, examine ChatGPT’s roles across the five inquiry phases, and analyze its advantages and challenges. The reviewed studies showed an uneven disciplinary distribution, with science receiving the most attention, whereas technology, the arts, engineering, and integrated STEAM remained underrepresented. The studies were evenly distributed between K–12 and higher education contexts; however, undergraduate students constituted the largest single participant group, and no study focused on elementary learners. ChatGPT was primarily used during the conceptualization, investigation, and discussion phases, where it functioned as a learning tool, tutor, learning peer, domain expert, and teaching assistant to support question formulation, inquiry design, problem-solving, and reflection. Its use in the orientation and conclusion phases remains largely unexplored. The studies reported several advantages of using ChatGPT. For students, it enhanced academic performance, critical thinking, engagement, and motivation, while also supporting individualized learning. For learning and teaching processes, it improved interaction, reduced educator workload, and facilitated differentiated learning environments, ultimately enhancing teaching efficiency. However, the studies also identified multifaceted challenges. At the learner level, students may over-rely on ChatGPT-generated answers, accept inaccurate or irrelevant feedback, experience conceptual confusion, or produce superficial conclusions when AI-generated outputs are treated as authoritative rather than critically evaluated. ChatGPT may also provide generic or insufficiently personalized responses that fail to stimulate divergent thinking. In addition, hallucinations, including inaccurate or fabricated explanations, may undermine evidence synthesis and reasoning during the conclusion phase. At the educator and institutional levels, challenges included misalignment with curricular goals, reduced instructional depth, the need for sustained teacher mediation, unequal access, technical barriers, privacy concerns, risks to academic integrity, and algorithmic bias. An integrated framework is proposed that synthesizes ChatGPT’s roles, advantages, and challenges across the five phases of inquiry-based learning, along with four agendas for future research. These findings offer guidance for responsible and pedagogically aligned integration of ChatGPT into STEAM inquiry-based learning.
Students’ STEM identity has received considerable attention in STEM education in recent years. However, the diversity in how it is operationalized and measured in quantitative studies has led to conceptual confusion and difficulties in comparing findings across studies. This study aims to organize and represent operationalized components of student STEM identity from quantitative measurement studies by applying standards and techniques from ontology engineering, a knowledge modeling approach that systematically represents concepts and their interrelationships in a structured, machine-readable form. We developed the Student STEM Identity Ontology (S-STEMIO), which organizes and integrates empirically derived components of student STEM identity from quantitative measurement studies into a coherent hierarchical framework. S-STEMIO clarifies conceptual boundaries, distinguishes overlaps, and establishes explicit relationships among components of student STEM identity, thereby offering a structured, machine-readable representation of how student STEM identity has been operationalized across the quantitative literature. Within the S-STEMIO, these components cluster into distinct ontological classes, reflecting different assumptions about whether identity is something students possess, enact, and/or develop over time. Three overarching definitional orientations toward identity were identified: trait-based/self-perception, situated/contextualized, and performative/behavioral. Students’ internal perceptions and values were found to be more frequently examined, whereas behavioral and developmental aspects were less explored. Relationships among components were defined by specifying their hierarchical structure and semantic relations, thereby linking dispositional, representational, and enacted aspects of identity within a coherent framework. S-STEMIO provides a unified framework for student STEM identity that brings conceptual clarity, facilitates communication among researchers, and supports measurement instrument development. Serving as a conceptual map of current quantitative STEM identity research, it not only shows where the field stands but also guides future research directions.
Abstract Background, context, and purpose Although funding agencies make large investments and encourage program adoption, little is known about the process of institutionalizing grant-funded efforts in higher education. This multiple case study examined how grant-funded awardees of a biomedical training grant worked toward institutionalizing program activities. Cases included 10 diverse postsecondary institutions (368 study participants) across the United States charged with implementing a federal grant aimed at comprehensive and multi-level approaches to engage and retain students from diverse backgrounds in biomedical research. Findings Employing prior frameworks on institutionalization in higher education, we examined how embedded campus agents appealed to two determinants of institutionalization: compatibility and mutualism. We identified institutional resources as an additional determinant. Compatibility is most commonly sought after as early as the grant proposal stage, while agents typically appealed to mutualism in the later stage of implementation. Institutional resources are sought throughout all stages of the grant. Within each determinant, we describe strategies and tools that embedded agents use to facilitate long-term institutionalization of grant-funded efforts. We also describe common challenges sites faced as they worked to obtain compatibility, mutualism, and resources. Conclusions Leaders of postsecondary science training initiatives who aim to institutionalize efforts can utilize a variety of strategies to enhance compatibility, mutually shared benefits, and committed resources to continue, embed, and/or scale the original initiative.
Abstract Background Latine students in the United States face persistent barriers to educational opportunities, especially in science, technology, engineering, and mathematics (STEM) fields. Although organized after-school activities hold promise as structural supports for adolescents marginalized in STEM, more research is needed to understand how to make them effective, empowering spaces for all students. Culturally responsive practices have been shown to be beneficial in classrooms and may offer a promising approach but have been understudied in after-school activities. Grounded in culturally relevant pedagogy and culturally responsive practices, this qualitative study explored students’ perceptions of culturally responsive practices within a university-based after-school math enrichment activity. Methods In winter and spring 2024, semi-structured qualitative interviews were conducted with 26 purposefully selected middle school students (58% girls, 100% Latine) from the larger population of 104 student participants of an after-school math enrichment activity. Reflexive thematic analyses explored student-identified culturally responsive learning practices at the after-school activity and how they compared to in-school learning practices. Results Based on deductive and inductive coding of the qualitative data, adolescents identified practices spanning all theoretical domains of culturally responsive practices as central to their learning in the after-school activity. Students described how these practices enhanced their confidence, persistence, and engagement in mathematics. Though some practices resembled those they experienced in school (e.g., collaborative problem solving), students emphasized that the after-school activity also provided unique opportunities—such as individualized guidance, youth-led approaches, and a reduced emphasis on grades—that made them feel more supported, and that the math they learned was relevant to their lives. Conclusions Findings provide empirical evidence that culturally responsive practices can be meaningfully enacted in after-school STEM activities and experienced positively by Latine youth. By affirming students’ cultural identities and leveraging their strengths, these practices helped create inclusive environments where adolescents viewed themselves as capable math learners. As after-school STEM activities continue to expand, embedding culturally responsive practices offers a promising pathway to broadening participation and supporting students’ persistence and success in STEM.
Students who believe intelligence is malleable rather than fixed – who hold a growth mindset – demonstrate increased motivation, resilience, and academic success. However, evidence for the effectiveness of mindset interventions in higher education is mixed, and most studies focus on introductory STEM courses and performance outcomes such as grades. Little is known about mindset effects in upper-level, research-intensive courses where students develop transferable scientific process skills critical for the STEM workforce. Engagement in authentic research practices is cognitively and emotionally demanding, and the impact of a growth mindset at this stage of the STEM pipeline remains unclear. We conducted a retrospective analysis of a simple, scalable growth mindset intervention embedded in research article coursework in an upper-level neuroscience course. The intervention consisted of a repeated reflective prompt encouraging students to interpret struggle as a normal, productive part of learning. Outcomes included students’ perceptions of research-related skill development, scientific self-efficacy, research community belonging, qualitative course feedback, and exam-based measures of figure analysis. Underrepresented minority (URM) and first-generation (FG) students in the intervention group showed significantly greater gains in research community belonging than control peers. FG students also reported greater gains in selected research-related skills, including scholarly writing and presenting research. No significant differences were observed in direct exam-based performance or overall self-efficacy. A brief growth mindset reflection integrated into authentic research coursework was associated with increased research community belonging among historically marginalized students in an advanced STEM context. While limited by its retrospective design and reliance primarily on self-report measures, this study extends mindset intervention research beyond introductory courses and performance metrics. It also provides a simple approach to enhance integration of research into undergraduate STEM education. Finally, our findings also suggest that belonging may be an important mechanism through which mindset-supportive practices promote persistence, particularly during research skill development. These results provide a foundation for prospective and longitudinal studies of mindset interventions aimed at supporting equity and persistence in STEM.
This systematic review examines model-based reasoning (MBR) in STEM education, focusing on how it is defined, how it works in practice, and how it is measured and taught across classrooms and laboratories. Guided by PRISMA 2020, this review synthesized 146 peer-reviewed studies published between 1980 and 2025 to address four research questions. We report a qualitative synthesis and descriptive frequencies from the findings. First, the literature converges on MBR as an iterative, distributed, and representation-mediated practice that links mental models with external inscriptions, including diagrams, equations, prototypes, code, and simulations, while integrating abductive, inductive, deductive, causal-mechanistic, and computational forms of reasoning. Second, the reviewed studies suggest recurring stage-based patterns in modeling and simulation activities: abductive and analogical reasoning are especially visible during early problem analysis and formulation; deductive, quantitative, and algorithmic reasoning support model construction and execution; diagnostic, inductive, and probabilistic reasoning support verification, validation, and debugging activities. Third, the field broadly agrees on the centrality of iteration, external representations, and collaboration in model-based reasoning activities, while debates persist over primary theoretical emphasis, particularly whether model-based reasoning is best grounded in mental models or distributed cognition; whether reasoning modes should be treated as analytically separable or as hybrid in use; and the extent to which domain-specific standards should guide model evaluation and acceptance decisions. Fourth, the ways MBR is characterized and measured shape what can be claimed about it: micro-level approaches, such as think-aloud protocols and time-stamped coding, capture moment-to-moment strategy use; meso-level approaches, such as computational notebooks, simulation logs, and rubric-based assessments, reveal workflow and representational competence; and macro-level approaches, such as model-evidence link diagrams and portfolios, capture longer-term development over weeks or semesters. These findings position MBR as a useful integrative lens for scientific sensemaking that is applicable across STEM disciplines, while also showing that its meaning and assessment remain shaped by disciplinary, instructional, and methodological context.
Abstract Background The STEMM sector is increasingly vital, with a growing demand for skilled professionals. Beyond cognitive skills, non-cognitive factors such as motivation, affect, and attitudes have a significant influence on STEMM performance and aspirations. To gain an overview of the field of research, a second-order meta-analysis was conducted on the effectiveness of interventions on non-cognitive outcomes in STEMM education. A systematic search (2000–2023) identified 6664 review studies, from which 42 meta-analyses containing 64 meta-analytic effect sizes were included. These encompassed 1903 primary studies, covering both K-12 and higher education contexts. The interventions were categorized by the targeted outcome, intervention method, and targeted STEMM discipline. Results Mathematics was the most frequently studied subject, and common interventions included applications of educational technology and student-centered instruction. The most studied outcomes were attitudes towards STEMM and affective factors, such as mathematics anxiety and satisfaction. Overall, interventions showed a significant positive effect on non-cognitive outcomes (d = 0.42, 95% CI [0.35, 0.50], p < .001), indicating improved motivation, attitudes, or reduced anxiety toward STEMM. Effectiveness did not differ significantly across outcome types or intervention methods, although mathematics and general STEM studies showed slightly smaller effects. We also found substantial variability in the meta-analytic effects. Notably, gaps in the literature were identified. Affective states such as boredom and enjoyment were rarely addressed, and out-of-school contexts were underrepresented. Conclusion This highlights the need for more comprehensive and diverse meta-analyses. Nonetheless, student-centered teaching, games, and technology-based methods appear promising for enhancing non-cognitive outcomes in STEMM education.
Abstract Educational research in scientific disciplines has gained increasing importance as a source of scholarly support for evidence-based improvements in science teaching and learning. Trends in this area can be examined not only through individual publications on specific topics over time, but also through the broader development of science-based education journals. To capture these trends, this study examines the inclusion and impact of education-focused journals indexed in Clarivate’s Science Citation Index Expanded over time. The findings suggest that the nature and scope of educational scholarship in scientific disciplines have gradually evolved, characterized by the increasing visibility and impact of multidisciplinary and interdisciplinary research in STEM and STEMM education.
STEM identity is widely linked to students’ pursuit of STEM pathways. However, prior research has focused largely on students’ current sense of belonging to STEM communities, paying less attention to identity construction processes. Drawing on Marcia’s identity status theory and extended models, STEM identity construction can be described through indicators such as commitment, in-depth exploration, practices, affirmation, and reconsideration of commitment. Several identity statuses such as achievement, foreclosure, moratorium, and diffusion have been categorized to reveal individuals’ envisioned possible selves in the future. Moreover, identity formation reflects the internalization of the complex interactions among personality characteristics (e.g., self-esteem), psychological needs (e.g., values, interests and competence), and social support (e.g., support from parents and peers). However, few research has tested whether STEM identity statuses differ systematically in psychological needs and social support. To address this gap, we identified STEM identity statuses among Chinese university students and examined how these statuses relate to psychological needs and perceived social support. Using affirmation, in-depth exploration, practice, commitment, and reconsideration of commitment scores from 235 Chinese college students, we conducted a cluster analysis and identified five STEM identity statuses: Achievement, Utilitarian Achievement, Foreclosure, Moratorium, and Diffusion. Subsequently, we further explored gender and STEM discipline differences across these identity statuses and found no significant differences. Multivariate analysis of variance further indicated that students in the Achievement and Utilitarian Achievement groups reported significantly higher levels of STEM motivational beliefs, STEM activities experiences, and parental and peer support than those in the Moratorium and Foreclosure groups, whereas in the Diffusion group scored the lowest across all these outcomes. Identity status theory offers a useful lens for understanding the construction of STEM identity and distinguishing several STEM identity statuses. Examining STEM identity statuses and their associations with STEM motivation beliefs, STEM activity experiences, and perceived social support would also be beneficial for deeply understanding college students who persist in or escape from the STEM field. Students in Achievement and Utilitarian Achievement statuses demonstrate stronger STEM motivation, greater engagement in STEM-related activities, and higher level of parental and peer support, making them more likely to persist in STEM fields. In contrast, students in the Moratorium and Diffusion statuses tend to show greater uncertainty, lower motivation, and limited engagement, indicating a higher risk of leaving STEM pathways. These findings suggest that STEM identity status may serve as a useful diagnostic tool for the early identification of students at risk of disengaging from STEM trajectories.
