## Slide 1

Open by saying that every value and participant in this demonstration is synthetic. The question matters because analysts often treat missing evidence as an inconvenience rather than part of the recommendation. Ask the audience to distinguish noticing uncertainty from making a better decision.

## Slide 2

Describe the narrow gap. Existing classroom practice is background context in this fictional brief, not a cited empirical claim. In a real defense, replace it with the supplied literature and add precise source identifiers. Keep transfer and feasibility separate.

## Slide 3

Explain why the comparison is paired and why the result remains descriptive. Participants see the same task domain; that improves comparability but limits transfer. Reviewers code uncertainty recognition rather than whether they agree with the final recommendation.

## Slide 4

Walk through inclusion, assignment, timing and preservation in order. The alternating assignment is reproducible but not claimed to be random. Explain the incomplete-pair exclusion before showing the findings, so the analysis population is not chosen after the result.

## Slide 5

Read the unit before each value. The difference is uncertainties per response, timing is seconds, and agreement is a raw percentage. None of the three quantities is a measure of business impact. Invite inspection of the synthetic appendix labels.

## Slide 6

Deliver the interpretation in one sentence and stop. More caveats could be thoughtful or performative; that ambiguity is why the claim is bounded. The proposed decision is a transfer pilot, not organization-wide adoption.

## Slide 7

Treat alternative explanations as tests of the argument rather than objections to hide. Order effects, demand characteristics and taxonomy learning make different predictions. A good next study should discriminate among them.

## Slide 8

State each limitation and its implication. Avoid the empty phrase more research is needed: explain exactly which inference each limitation prevents. There are no medical or financial application claims in this deck.

## Slide 9

Separate the classroom use from evidence about effectiveness. A reflective exercise can be useful without being validated as a decision intervention. Never circulate the synthetic values as observed participant data.

## Slide 10

Use these prompts to rehearse concise answers. Answer what was measured, name what was not, and point to the appropriate appendix. If a question requires missing evidence, say so and record the follow-up.

## Slide 11

Explain the discriminating purpose of the next test. A neutral instruction addresses demand effects, blinded decision scoring addresses the endpoint gap, and preregistration limits retrospective analytic choices.

## Slide 12

Keep the fictional source register visible for questions. For real work, replace every locator with an accessible artifact or citation that the researcher actually supplied. Do not invent a DOI, participant count or unpublished result.