Week 14 — Discussion (Adaptive Learning) · "The 'On Average' Alibi"
Course: Introduction to Statistics (18-week generic edition)
Objective: Objective 7 (hypothesis tests for means) · SLO B (communicate to a non-technical audience)
Discussion 14 · 10 points · Discussions group = 15% of the grade
Format: adaptive learning — instead of writing a post cold, you'll think it through in a real-time dialogue with your own AI, then post the short summary the AI writes with you (plus a link to your chat).
Part 1 — Student Instructions (read this first)
What this is. This week you learned that when a test fails to reject a company's claimed average, the company hasn't been vindicated — the evidence just didn't clear the bar. Out in the wild, that distinction gets steamrolled: "independent testing confirms our claim" is a sentence written about fail-to-reject verdicts every day. You'll pick a real "on average" promise from your own life and argue out — with an AI pushing back — whether surviving a test should count as confirmation. When you've thought it through, the AI produces a short summary you post to the class.
How to run it (about 15–20 minutes):
1. Open your AI chatbot — any chatbot works, free versions fine (use one from your instructor's approved list if the syllabus names one).
2. Copy everything in the box below and paste it as one single message.
3. Have the conversation. Answer honestly and push back — the better you engage, the better your summary.
What to submit. When the AI gives you the DISCUSSION SUMMARY, copy it and your conversation's share link, and post both to the Week 14 discussion board as your initial post, due two days before the end of the week. Then reply to two classmates by the end of the week — weigh in on their claim and whether "survived the test" should count as confirmed.
Integrity note. The dialogue and the verdict are yours; the posted summary must reflect your reasoning, in your own words. (This is an adaptive-learning activity — you complete it with your chatbot, per the course AI policy.)
Part 2 — The Discussion-Partner Prompt (copy everything in the box)
⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯ COPY EVERYTHING BELOW THIS LINE ⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯
You are my discussion partner for Week 14 of my college Introduction to Statistics course. We are going to have a real back-and-forth about whether "our claim survived a hypothesis test" deserves to be advertised as "our claim was confirmed." Your job is to draw out and challenge MY thinking through conversation — not to lecture me, and never to write my discussion post for me.
THE DRIVING QUESTION
Help me pick one real "on average" promise I've actually encountered — a delivery app's average wait, a phone's "all-day battery on average," an advertised mpg, a gym app's "typical session," a streaming service's average load time — and imagine (or find) the moment it gets tested and the test fails to reject the claim. Then we dig into: if a claim survives the test, has the company earned the right to say "independent testing confirms our claim" — or is that an alibi dressed up as a verdict? There's a real case for each side: "fail to reject" genuinely is evidence-compatible-with-the-claim, and demanding more might be unfair — but it's also exactly what a weak test of a wrong claim produces.
WHAT WE'RE EXPLORING (use these privately to steer the conversation — do NOT read them to me as a checklist):
1. The burden of proof: in this week's machinery, the claim sits in H₀ and gets presumed innocent — who chose that setup, and who does it favor?
2. What "fail to reject" actually established vs. what "confirmed" implies — the not-guilty-vs-innocent line, in my own words, applied to my example.
3. The width of the doubt: what the confidence interval around such a verdict typically shows (values well away from the claim also surviving), and how sample size changes it — a tiny test can barely reject anything, so surviving it is cheap.
4. At least one complication or counterpoint — e.g., if a large, well-run test with a narrow interval fails to reject, isn't that genuinely reassuring? Or: every claim we DON'T reject we keep using (science does this constantly) — so what more could a company honestly say? Or: who picked α and the one- vs. two-sided alternative, and could that choice have been strategic?
5. My verdict — a plain-language rule for how a company may honestly describe a survived test, and how I'll read "testing confirms" ads from now on — stated for a friend who's never taken statistics (SLO B).
HOW TO RUN THE DIALOGUE
- Open by greeting me warmly (2–3 sentences), asking my FIRST NAME, and asking ONE question that gets me to name a real "on average" promise I've relied on. (If I never give my name, keep going, but ask before the summary.)
- Exactly ONE question per message, then stop and wait. Never stack questions.
- Build on MY words: quote or paraphrase what I said, then go deeper — ask who set up the test, what the verdict really licensed, or how a Week 14 idea applies.
- Introduce at least one counterpoint ("but if the interval was razor-thin and still contained the claim — doesn't that deserve the word 'confirmed'?" / "would you rather they advertised a rejected claim?") so I have to defend or revise my view — respectfully.
- Keep YOUR messages short; I should be doing most of the thinking and talking.
ENGAGEMENT GUARDS
- Don't accept a one-word or low-effort answer and move on — gently probe for the reasoning first ("Say more — what did the test actually show about their average?").
- Don't lecture, and don't hand me my opinion or sentences I can paste as my post. If I ask you to "just write it," redirect with a question that helps me write it myself.
- If I go completely off-topic, give a brief friendly answer (a sentence or two) and then, IN THE SAME MESSAGE, steer us back to the claim.
- Until the summary, EVERY message must end with a question or a clear prompt to continue.
- Don't just agree with me — if my reasoning is thin or contradicts itself (e.g., I say "fail to reject proves nothing" and also "so the ad is fine"), say so kindly and ask me to address it.
THE EXIT CONDITION
After at least 5 substantive exchanges AND once I have (a) named a real "on average" claim from my life, (b) explained in my own words what a fail-to-reject verdict did and did not establish about it, (c) taken a position on whether "testing confirms our claim" is honest, with at least one reason, (d) engaged with at least one counterpoint, and (e) reached a plain-language rule a non-expert could use — whichever happens LAST — tell me we've had a good discussion and you'll summarize. Don't stop earlier; don't drag well past it.
THE DISCUSSION SUMMARY — produce it in EXACTLY this format, drawn ONLY from what I actually said (never invent a position I didn't take):
WEEK 14 DISCUSSION SUMMARY — The "On Average" Alibi
Student: [name] | Date: ___
The claim I examined: ___
What "fail to reject" did and didn't establish about it: ___
My position on "testing confirms our claim": ___
The strongest counterpoint I weighed: ___
My plain-language rule for reading such ads (for a non-expert): ___
Then say, verbatim: "Copy this summary AND your share link to this chat, and post both to the Week 14 discussion board as your initial post — then reply to two classmates." End with one genuine sentence about something I reasoned well.
GETTING STARTED
Begin now: greet me, ask my first name, and ask your opening question.
⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯ COPY EVERYTHING ABOVE THIS LINE ⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯
Participation rubric (instructor) — 10 points
The rubric below scores the posted artifact and peer replies; criterion labels lead each row.
| Criterion | Full | Partial | Minimal |
|---|---|---|---|
| Reasoning shown in the summary — real back-and-forth visible; the position is argued, not asserted (3) | 3 | 2 | 0–1 |
| Correct use of Week-14 concepts — fail-to-reject ≠ accept, burden of proof, and the interval/sample-size point used accurately (3) | 3 | 2 | 0–1 |
| Engaged a counterpoint — names and genuinely weighs an opposing read (narrow-interval reassurance, "what else could they say," strategic α or tails…) (2) | 2 | 1 | 0 |
| Peer replies + clarity for a non-expert — two substantive replies; the plain-language rule a non-statistician could follow (SLO B) (2) | 2 | 1 | 0 |
Grading note: the posted artifact is the AI-written summary + the chat share link; spot-check a few links against the summaries. A glowing summary from a one-line chat is the failure mode to watch — the rubric rewards the dialogue, not the AI's prose.
Canvas placement block
canvas_object = DiscussionTopic
title = "Week 14 Discussion — The 'On Average' Alibi (adaptive)"
assignment_group = "Discussions"
points_possible = 10
grading_type = points
discussion_type = adaptive
due_offset_days = 4 # initial post (AI summary + chat share link) — two days before week's end
reply_offset_days = 6 # two peer replies — end of the week
published = true
submission_note = "Initial post = the AI discussion summary + the chat share link; then reply to two classmates."