Facilitator Toolkit
The facilitator toolkit provides practical methods for running fair, focused, and productive Credencing discussions.
Before the Meeting
Choose one claim, one concept, and one case. A group that tries to discuss “truth,” “politics,” and “AI” all at once will usually drift into slogans.
Prepare a visible board with four columns: Claim, Evidence, Confidence, Update Condition. This keeps discussion concrete.
During the Meeting
Invite participants to state confidence levels in ranges, such as 40-60 percent, rather than demanding exact numbers. Ask follow-up questions gently: what evidence moves you, what evidence would lower confidence, and what uncertainty remains?
When conflict rises, slow the conversation. Ask each side to restate the other side’s strongest evidence before replying.
After the Meeting
Close by asking each person to write one sentence: “My confidence changed from ___ to ___ because ___.” If confidence did not change, ask what evidence would have changed it.
Groups improve when they track updates over time. A simple prediction log or discussion journal can turn conversation into calibration practice.
Feedback Language
Useful feedback is specific and non-accusatory. Try: “I hear strong confidence, but I am not yet hearing the evidence that supports that level.” Or: “It sounds like the evidence changed, but the confidence did not. Is that right?”
Avoid turning the model into labels for people. Diagnose belief states, not whole persons.
Discussion Questions
Questions with expandable guidance
Open each question for facilitator notes, examples, scenarios, and feedback. The aim is careful discussion at a senior high school reading level.