教育学最新文献

Introduction: Professional identity formation (PIF) is increasingly understood as a socially situated workplace learning process through which medical students develop identities and practices of physicians. Through participation in clinical work within the clinical learning environment (CLE), students learn what it means to become members of the profession. Within this context, a sense of belonging may be an important mechanism linking the CLE to PIF. However, quantitative evidence examining these relationships remains limited. This study examined whether sense of belonging mediates the relationship between the CLE and PIF.

Methods: This cross-sectional study used data from 459 undergraduate clinical students in Indonesia collected between November 1 and December 15, 2024. Students completed three validated instruments: the Professional Identity Formation (PIF) questionnaire, the Undergraduate Clinical Education Environment Measure (UCEEM), and the Belongingness Scale-Clinical Placement Experience (BES-CPE). Structural equation modelling (SEM) was employed to assess whether sense of belonging functioned as a mediator between the CLE and PIF.

Results: PIF showed a moderate positive correlation with both the CLE (r = 0.416; p < 0.001) and sense of belonging (r = 0.370; p < 0.001), while the CLE was strongly correlated to sense belonging (r = 0.578; p < 0.001). The SEM mediation model demonstrated good fit (SRMR = 0.038, CFI = 0.96, TLI = 0.94), indicating that belonging partially mediated the relationship between the CLE and PIF.

Discussion: The findings reveal that PIF are positively linked to both CLE and a sense of belonging. In addition, sense of belonging functions as a significant statistical mediator in the pathway linking the CLE to PIF. These findings suggest that workplace learning experiences contribute to PIF not only directly but also through fostering students' sense of belonging. Creating supportive clinical environments that promote participation, inclusion, and membership may therefore strengthen medical students' PIF.

Objective Structured Clinical Examinations (OSCEs) remain central to the assessment of clinical competence, yet the integration of artificial intelligence (AI) into assessment design, delivery, and scoring raises a fundamental question: what evidence of competence are OSCEs intended to generate? This paper argues that AI should be evaluated through the lens of OSCE validity, with particular attention to how it may strengthen or weaken the interpretive chain linking observed performance to assessment decisions. Drawing on Messick's unitary conception of validity and Kane's argument-based framework, we examine how AI may introduce distinct threats across the scoring, generalisation, extrapolation, and implication inferences that underpin OSCE interpretation. We argue that generative AI (genAI) creates a particular risk of construct drift, whereby plausible assessment materials or scoring processes gradually reward observable proxies rather than the underlying clinical capabilities they are intended to assess. To translate this validity perspective into practice, we propose a framework that maps AI applications across the OSCE workflow according to their proximity to candidate consequences and the evidential burden required to justify their use. Lower-risk applications, such as design support and post-hoc analytics, may strengthen assessment when accompanied by construct-focused review and human oversight, whereas higher-risk uses, including AI-mediated candidate encounters and automated summative scoring, require substantially stronger validity evidence. We conclude that AI should be adopted according to the principle of concept before capability: assessment communities must first establish what clinical competence means and what evidence should count before deciding where AI has a legitimate role in high-stakes assessment.

Graduate employability is a priority for many sectors of the community, including higher education, government, and the students themselves. However, science graduates have gaps in identifying skills gained from their undergraduate degree and what they see as important for their chosen career. Therefore, it's important that curriculum and teams supporting student learning provide explicit opportunities for students to build awareness of their learned skills relevant to employment that are developed within curricula. This study aimed to develop a science employability framework to help undergraduate science students identify and understand the skills they gained from their studies and their relevance to their future careers. Firstly, analysis was conducted to determine if employability skills are embedded into current curricula at the study site. Secondly, the currency of these skills was evaluated by a body of work to identify newly articulated skills in recent position descriptions for jobs targeted by the courses. These analyses were completed using tools developed using Natural Language Processing (NLP). The skills from these two bodies of work were then organized into a three-pillar framework (professional, knowledge, and technical skills) for common relevant positions. Using the employability framework, we made "Employability Cards" which were implemented into an interactive workshop activity designed to enhance students' awareness of job opportunities relevant to their degree. Pre- and post-workshop surveys were conducted on participating students to determine the current knowledge and perceptions on employability.

This study investigates assessment of Specific Learning Disorders (SLD) in Twice-Exceptional (TE) children, gifted children (GC) who also present SLD, in oral and written language domains. Applying the standard threshold for identification (< 7th percentile) to GC often results in the assessment and identification of SLD in the TE population, thereby limiting access to appropriate support. The objective was to determine which language tests effectively identify SLD in TE children and establish optimal threshold values. The prospective study included 301 children 7–16 years, divided into three groups: GC without SLD (n = 71), TE children (n = 114), and non-gifted children with SLD (n = 116). Language test performances were compared across groups, and Classification and Regression Tree (CART) analyses were employed to optimize test sensitivity and specificity in the TE group. TE children performed better than the SLD group but worse than GC without disorders. At standard thresholds, text reading and spelling tests showed the highest sensitivity, which significantly improved when CART-derived decision thresholds (62nd and 38th percentiles) were applied, while maintaining satisfactory specificity. Results indicate that standard thresholds lead to under-identification of SLD in GC, and that adjusted thresholds enhance identification accuracy. These findings should generalize to similarly complex languages, such as English.