<p>Evidence-informed practice (EIP) in instructional design is frequently constrained by epistemic and structural barriers that impede the integration of scientific research into instructional design practitioners’ workflows. Building upon the foundational distinctions established in Part I, this article acts as Part II and proposes a pragmatic framework for operationalizing EIP within real-world design constraints. We introduce a systematic workflow grounded in the Learner, Intervention, Context, Outcomes (LICO) framework to refine inquiry and inclusion criteria. To meet the demands of professional practice, we advocate rapid reviews as a rigorous yet agile synthesis method. Furthermore, we examine how artificial intelligence can augment search, screening, and drafting processes, while emphasizing the need for human verification to ensure reproducibility. For evidence appraisal, we propose a Red–Amber–Green (RAG) framework to assess study quality and risk of bias. This position paper offers an actionable methodology for designers to acquire and apply scientific evidence, with broader implications for professional development.</p>

错误:搜索内容不能为空,请输入英文关键词
错误:关键词超出字数限制,请精简
高级检索

AI-Assisted Evidence-Informed Practice: Reclaiming the Scientific Foundations of Instructional Design-Part II (How)

  • Atsusi Hirumi,
  • Lisa Giacumo,
  • Dina Kurzweil,
  • Efren de la Mora Velasco,
  • Henry Moon

摘要

Evidence-informed practice (EIP) in instructional design is frequently constrained by epistemic and structural barriers that impede the integration of scientific research into instructional design practitioners’ workflows. Building upon the foundational distinctions established in Part I, this article acts as Part II and proposes a pragmatic framework for operationalizing EIP within real-world design constraints. We introduce a systematic workflow grounded in the Learner, Intervention, Context, Outcomes (LICO) framework to refine inquiry and inclusion criteria. To meet the demands of professional practice, we advocate rapid reviews as a rigorous yet agile synthesis method. Furthermore, we examine how artificial intelligence can augment search, screening, and drafting processes, while emphasizing the need for human verification to ensure reproducibility. For evidence appraisal, we propose a Red–Amber–Green (RAG) framework to assess study quality and risk of bias. This position paper offers an actionable methodology for designers to acquire and apply scientific evidence, with broader implications for professional development.