Grab lunch and join our sponsors to learn more about their products, innovations, and thought leadership in the field. Lunch and learners have a limited capacity and participation is on a first-come, first-served basis.
Center for Research Evaluation (CERE) at the University of Mississippi
Reading the Terrain: Situation Awareness as a Tool for Engaging Evaluation Clients
Thursday, November 19
12:45 p.m. – 1:45 p.m.
Room: E220E
Speakers: Sarah Mason and Rachel Scott
Effective evaluation requires understanding and responding to context. Yet evaluation involves navigating intersecting terrains: relational (who are we working with?), organizational (what cultures, structures, incentives, and power dynamics are at play?), boundary (how do we differ from the people and places we evaluate?), and dynamic (how is the situation changing?).
At the Center for Research Evaluation (CERE) at the University of Mississippi, we use the Framework for Situation Awareness as a practical tool for reading and navigating these terrains.
In this Lunch and Learn, CERE will introduce the Framework and explore how evaluators can use it to understand and adapt to context, build lasting client relationships, and design evaluations grounded in shared understanding.
Evidano
What Evaluators Need to Know About AI in 2027
Friday, November 20
12:15 p.m. – 1:15 p.m.
Room: E212ABCD
Speaker: James Goh, Founder of Evidano
As we approach 2027, AI in evaluation presents an interesting contrast: remarkable advances in what is possible, yet surprisingly persistent challenges around reliability, nuance, and ethical judgment.
AI has evolved well beyond ChatGPT. Hallucinations are becoming more manageable, and tasks once considered too complex or unreliable for AI, such as analyzing 100+ documents, coding qualitative data, identifying nuanced themes, and synthesizing multilingual evidence, are becoming increasingly practical. Some AI-assisted qualitative analysis approaches have already demonstrated greater than 90% accuracy relative to human experts in independent peer-reviewed research.
Yet significant limitations remain. AI still struggles with cultural context, ambiguity, minority perspectives, causal reasoning, and complex evaluative judgments. Even as technical accuracy improves, important questions persist around bias, confidentiality, transparency, accountability, and whether greater efficiency comes at the expense of methodological rigor.
This practical Lunch & Learn will examine what's changing from 2026 to 2027, drawing on real evaluation workflows and case studies involving organizations including UNICEF, UNFPA, and UNESCO. Participants will explore where AI has made meaningful progress, where familiar problems persist, and how emerging capabilities create new ethical and methodological considerations.
Attendees will leave with a clearer understanding of what AI can now do, what it still cannot do reliably, and how to balance technological advances with the nuance, ethics, and human judgment essential to good evaluation.