This conceptual paper proposes a framework for redesigning asynchronous assessments in online graduate nursing education in response to generative artificial intelligence, and proposes four strategies grounded in the community of inquiry framework, authentic assessment theory, experiential learning theory, and transformative learning theory: experience-anchored analysis, multimodal documentation portfolios, asynchronous verbal exchange, and asynchronous collaborative process.
Abstract
This conceptual paper proposes a framework for redesigning asynchronous assessments in online graduate nursing education in response to generative artificial intelligence (AI). Traditional text-based assignments are vulnerable to AI-generated submissions, and conventional plagiarism detection software performs poorly against such content. We argue for redesigning assessments, rather than increasing surveillance. We propose four strategies grounded in the community of inquiry framework, authentic assessment theory, experiential learning theory, and transformative learning theory: experience-anchored analysis, multimodal documentation portfolios, asynchronous verbal exchange, and asynchronous collaborative process. None of the strategies are new, and each has a long history in nursing and health professions education. The contribution of this work is its integration of the four strategies into a single design rationale that would be responsive to the pressure placed by AI on asynchronous assessments. We use ‘AI-resistant’ in this paper in a bounded sense, that is, for designs that can reduce the feasibility of the substitution of students’ work with AI, rather than designs that can prevent any form of AI involvement.
Embedding structural changes within assessment design, rather than relying on rule enforcement, will ensure nursing and midwifery graduates are prepared to thrive in an AI-enabled world.
Y. Salamonson, Pauletta Irwin, R. Kornhaber et al.· Journal of Clinical Nursing· 0 citations
This commentary is deliberately framed as a focused extension of the original study’s individual-level findings to the institutional level, rather than a stand-alone organizational literature review, and its scope has been kept correspondingly concise.
Preliminary findings suggest AI-enhanced case-based platforms can engage students and support applied ethics learning but are best positioned to complement rather than replace traditional instruction.
L. Rankine, J. Van Bussel, S. Moodie et al.· medRxiv· 0 citations
A framework for AI-resilient assessment that shifts evaluation from product quality to demonstrable reasoning, decision-making, and ownership of learning is proposed, Illustrated primarily through health sciences education, with wider relevance to professional and practice-oriented disciplines.
Dragan Nikolić, M. Basta Nikolić· Frontiers in Artificial Inte...· 0 citations
Background Artificial intelligence (AI) is quickly revolutionizing higher education; however, there is high heterogeneity regarding the adoption of AI amongst the students. In the context of nursing education, where successful learning, critical thinking skills, and professional accountability are crucial, it becomes essential to understand how the students interact with the AI applications. The current study was conducted to determine the AI-use profiles among nursing students and how they relate to academic outcomes and demographics. Methods We used a cross-sectional survey among nursing students. AI use for learning purposes was assessed using four binary indicators capturing use for understanding concepts, summarizing content, drafting assignments, and language support. Study engagement was measured using selected items from the Online Student Engagement framework and collapsed into three ordinal categories. We assessed academic performance using three self-reported items. We performed a latent class analysis to identify distinct student profiles based on AI use and engagement indicators. Multinomial logistic regression examined associations between demographic factors and class membership. Multiple linear regression assessed differences in academic performance across classes. Results The best model was a three-profile solution with a good classification accuracy. The profiles included strategic engagers, moderate users, and passive or low engagers. As expected, strategic engagers were characterized by high levels of engagement with all kinds of behaviors studied, whereas Passive used consistently reported low levels of engagement. Usage indicators of AI revealed little variation among classes. The performance scores of strategic engagers were significantly better compared with those of passive users (β = 2.39, 95% CI = 2.08–2.69). No significant association was found for demographic characteristics. Conclusions Nursing students can be categorized into distinct profiles based on their patterns of AI use and study engagement. Our findings showed that study engagement, but not AI use alone, was the key factor associated with academic performance. These findings highlight the importance of promoting effective learning strategies alongside AI integration. Educational interventions should focus on guiding students toward strategic and responsible use of AI to enhance learning outcomes.
Majed M. Aljabri, Bandar S. Alharbi, Endale Alemayehu Ali· Frontiers in Medicine· 0 citations