AI and (Empowerment) Evaluation: The AI Revolution Will Not be Televised (It is Live)

Abstract Information: AI and Evaluation have attracted a significant amount of attention, ranging from standing-room-only AEA sessions and an AI and Evaluation eStudy to the establishment of an AI and Evaluation LinkedIn page and the publication of a New Directions for Evaluation volume, dedicated to the topic. This session is designed to build on that thirst for AI and Evaluation knowledge. The session will begin with the “when, where, and flow” of AI and Evaluation. It will be followed by a series of applications of AI to evaluation, including helping evaluators develop responsive proposals, relevant program logic models, useful survey questions, and efficient analyses of data. The session will demonstrate how to use AI to conduct qualitative analysis, including high-level thematic coding of large datasets as well as open-ended responses and interviews. It will also highlight the development of evaluation GPTs, including James Penn’s Survey Question GPT and David Fetterman’s EmpowerEval Assistant GPT. The session will be focused on helping evaluators help themselves (and thus an empowerment focus). It will conclude with an invitation for participants to share their applications of AI to evaluation.