From Algorithms to Evidence: Using GenAI in Evaluation Practice
暫譯: 從演算法到證據：在評估實踐中使用 GenAI
Bruce, Kerry, Gandhi, Valentine J., Bohni Nielsen, Steffen
相關主題
商品描述
From Algorithms to Evidence: Using GenAI in Evaluation Practice offers a timely, practice-grounded guide for evaluators and development professionals navigating the fast-moving world of generative AI. Building on the foundations laid in Artificial Intelligence and Evaluation (2025), this volume moves decisively from theory to application. It documents how evaluators across the globe are already using GenAI in real projects, showing not only what worked, but also what failed, why, and under what conditions. The result is a clear, practitioner-centered resource that cuts through hype and provides grounded, real-world insight.
Structured around the seven phases of the evaluation lifecycle (design, structuring and inception, data collection, data analysis, reporting, judgment, and utilization) the book offers concrete examples of how GenAI is being integrated into everyday evaluation tasks. Contributors explain their rationale for using AI tools, the steps they took, and the results they achieved. Crosscutting chapters synthesize lessons on methodological adaptation, evolving evaluator competencies, and the ethical and professional standards needed to use GenAI responsibly. Throughout, the volume emphasizes "hybrid intelligence," showing how human expertise and AI-enabled methods can work together to strengthen evaluative reasoning.
Clear, accessible, and grounded in real practice, From Algorithms to Evidence fills a critical gap in the literature. It provides evaluators, policymakers, and organizational leaders with practical guidance for building more adaptive, data-informed, and future-ready evaluation systems, while keeping equity, transparency, and human judgment at the center of their work.
商品描述(中文翻譯)
《從演算法到證據：在評估實踐中使用 GenAI》提供了一本及時且以實踐為基礎的指南，幫助評估者和開發專業人士在快速變化的生成式 AI 世界中導航。這本書建立在《人工智慧與評估》（2025）所奠定的基礎上，果斷地從理論轉向應用。它記錄了全球評估者如何在實際項目中使用 GenAI，不僅展示了成功的案例，還分析了失敗的原因及其條件。最終，這本書成為一個清晰且以實務為中心的資源，穿透炒作，提供扎實的現實洞見。
本書圍繞評估生命週期的七個階段（設計、結構與啟動、數據收集、數據分析、報告、判斷和利用）進行結構化，提供了 GenAI 如何融入日常評估任務的具體範例。貢獻者解釋了使用 AI 工具的理由、所採取的步驟以及所取得的結果。跨章節的內容綜合了方法適應、評估者能力的演變，以及負責任使用 GenAI 所需的倫理和專業標準的教訓。整體而言，本書強調「混合智慧」，展示了人類專業知識與 AI 驅動方法如何協同工作，以增強評估推理。
《從演算法到證據》清晰、易讀且根植於實際操作，填補了文獻中的一個關鍵空白。它為評估者、政策制定者和組織領導者提供了實用的指導，幫助他們建立更具適應性、數據驅動和未來準備的評估系統，同時將公平、透明和人類判斷置於工作中心。
作者簡介
Kerry Bruce (DrPH) is a global health and evaluation practitioner working to support both practical monitoring, evaluation, and learning projects and training. She is the CEO of ClearUp Consulting and is a member of INTEVAL.
Valentine Joseph Gandhi (PhD) wears multi-disciplinary hats as a Policy Advisor, MEL specialist, and AI, Data Science and Cybersecurity expert. He is the founder of the think tank The Development CAFE, the host of the EvalEdge Podcast for the European Evaluation Society and a member of INTEVAL. Steffen Bohni Nielsen (PhD) is Director General at the Danish National Research Centre for the Working Environment (NFA), a member of the Danish National Research and Innovation Council, and chairman of the Partnership for European Research in Occupational Safety and Health (PERSOH). He is also a member of INTEVAL and has published extensively in the field of evaluation.作者簡介(中文翻譯)
Kerry Bruce（DrPH）是一位全球健康與評估實務專家，致力於支持實務監測、評估和學習專案及培訓。她是ClearUp Consulting的執行長，並且是INTEVAL的成員。
Valentine Joseph Gandhi（PhD）擔任多重職務，包括政策顧問、MEL專家，以及人工智慧、數據科學和網絡安全專家。他是智庫The Development CAFE的創辦人，也是歐洲評估協會EvalEdge Podcast的主持人，並且是INTEVAL的成員。
Steffen Bohni Nielsen（PhD）是丹麥國家工作環境研究中心（NFA）的總幹事，丹麥國家研究與創新委員會的成員，以及歐洲職業安全與健康研究夥伴關係（PERSOH）的主席。他也是INTEVAL的成員，並在評估領域發表了大量的研究。
