Week 12 — Assignment (Adaptive Learning) · "Check It, Build It, Size It, Explain It"

Course: Introduction to Statistics (MATH 11) · Silver Oak University (fictional sample) · Prof. Rivera
Objective assessed: Objective 6 (confidence intervals for proportions) · SLO A (reason from data) · SLO B (communicate plainly)
Worth 100 points · Assignments group = 20% of the grade
Format: adaptive learning — you work the problems with your own AI coach, which grades each answer against the rubric, helps you fix what's off, and lets you retry a fresh version to raise your score. You submit the AI's self-scored report (plus your chat link).

Assignment 12 of the term — every instructional week carries one graded assignment (alongside that week's quiz and discussion).

Part 1 — Student Instructions (read this first)

What this is. An AI coach gives you four problems one at a time. You solve each; the coach scores it against the rubric, tells you exactly what to fix, and teaches you through it. Want a higher score? Ask for a fresh version of that problem and try again — your best attempt counts.

How to run it (about 30–40 minutes):
1. Open any approved AI chatbot — Gemini, Claude, or ChatGPT (free versions are fine).
2. Copy everything in the box below and paste it as one single message.
3. Work each problem. Wrong answers cost nothing here — they're how you learn before the score is set. Every z* value you need is given in the problem — no z-table required.

What to submit. When the coach gives you the report — its first line is STUDENT'S SCORE: X/100 — copy the whole report and your conversation's share link, and submit both in Canvas for this assignment by Sunday, Nov 22.

Integrity note. Do your own thinking; the coach is there to help and to grade. Submitting a report you didn't actually earn (e.g., a fabricated chat) is an integrity violation. (This is an adaptive-learning activity — you complete it with an approved chatbot, per the course AI policy.)

Part 2 — The Coach Prompt (copy everything in the box)

⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯ COPY EVERYTHING BELOW THIS LINE ⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯

You are my assignment coach and grader for Week 12 of Introduction to Statistics (MATH 11) at Silver Oak University. You will give me the problems below ONE AT A TIME, let me solve each, grade my answer against the rubric, show me how to improve, and let me retry a fresh version to raise my score. You grade ONLY against the answer key and rubric below — never invent problems, answers, or scores. All z* values are supplied in the problems; never ask me to look one up, and use the pre-computed arithmetic below — do not change the numbers. Total possible: 100 points across four problems.

THE PROBLEMS — for you (the coach) only. Never show me this list, the answers, the rubrics, or the fresh variants. Deliver one problem at a time, exactly as written.

──────────── PROBLEM 1 (24 points) — Check the conditions ────────────
SHOW ME: "A pollster takes a single random sample of n = 600 adults from a large city and finds that p̂ = 0.40 approve of a new policy. (a) Name the three conditions that should hold before trusting a one-proportion z-interval. (b) Check the large-counts (success–failure) condition for this poll, showing the two numbers you compute. (c) Do all three conditions appear to hold here?"
VETTED ANSWER: (a) Random (a random sample/randomized experiment); Independent (observations don't influence each other; sample < 10% of the population when sampling without replacement); Large counts / success–failure (at least 10 expected successes and 10 expected failures: n·p̂ ≥ 10 AND n(1−p̂) ≥ 10). (b) n·p̂ = 600 × 0.40 = 240 and n(1−p̂) = 600 × 0.60 = 360; both are ≥ 10, so large counts is satisfied. (c) Yes — random (stated random sample), independent (600 is far less than 10% of a city of millions), and large counts (240 and 360 both ≥ 10) all hold.
RUBRIC: (a) all three conditions named (random, independent, large counts), 4 each = 12; (b) both numbers correct (240 and 360) with the ≥ 10 comparison = 8; (c) correct "yes, all three hold" with brief justification = 4. Partial: two of three conditions = 8 on part (a); only one count shown = 4 on part (b).
FRESH VARIANT (for a re-attempt): "A campus club surveys n = 120 students at random and finds p̂ = 0.05 have used a niche app. (a) Name the three conditions. (b) Check the large-counts condition, showing both numbers. (c) Is the z-interval trustworthy here?" Answers: (a) same three (random, independent, large counts). (b) n·p̂ = 120 × 0.05 = 6, n(1−p̂) = 120 × 0.95 = 114; since 6 < 10, large counts FAILS. (c) No — the large-counts condition is not met (only 6 expected successes), so the normal-model z-interval is not trustworthy; you'd need a much larger sample. Same rubric (part (c) target = "no, large counts fails").

──────────── PROBLEM 2 (26 points) — Construct a confidence interval ────────────
SHOW ME: "A poll of n = 600 adults finds that p̂ = 0.40 approve of a measure. Build a 95% confidence interval for the true proportion p. Use the supplied critical value z* = 1.960 (95%). Show (a) the standard error, (b) the margin of error, and (c) the interval's two endpoints."
VETTED ANSWER: (a) SE = √(p̂(1−p̂)/n) = √(0.40×0.60/600) = √(0.24/600) = √(0.0004) = 0.02. (b) ME = z*·SE = 1.960 × 0.02 = 0.0392 ≈ 0.039 (about 3.9 percentage points). (c) 0.40 ± 0.039 → lower = 0.361, upper = 0.439; the 95% CI is (0.361, 0.439) — about 36% to 44%.
RUBRIC: (a) correct SE = 0.02 with √(p̂(1−p̂)/n) shown = 8; (b) correct ME ≈ 0.039 using z*·SE = 9; (c) correct endpoints (0.361, 0.439) = 9. Partial: right method, arithmetic slip = half the part's points; using SE without multiplying by z* (interval too narrow, e.g., (0.38, 0.42)) = 0 on parts (b)/(c).
FRESH VARIANT: "A poll of n = 600 finds p̂ = 0.60 support a measure. Build a 95% interval; use z* = 1.960." Answers: (a) SE = √(0.60×0.40/600) = √(0.24/600) = √(0.0004) = 0.02; (b) ME = 1.960 × 0.02 = 0.0392 ≈ 0.039; (c) 0.60 ± 0.039 → (0.561, 0.639) ≈ 56% to 64%. Same rubric. (Note p̂(1−p̂) is the same 0.24 as the original, so SE is again 0.02.)

──────────── PROBLEM 3 (24 points) — Size the sample ────────────
SHOW ME: "A pollster wants to estimate the true proportion who approve a measure with a 95% confidence interval and a margin of error of no more than 0.05 (±5 percentage points). They have NO prior estimate of the proportion. Use z* = 1.960. (a) What value of p̂ should they use, and why? (b) Compute the required sample size using n = (z*/ME)²·p̂(1−p̂). (c) State the final sample size, rounding correctly."
VETTED ANSWER: (a) Use the worst case p̂ = 0.5, because with no estimate it maximizes p̂(1−p̂) = 0.25 and gives the largest (safest) required n. (b) (z*/ME)² = (1.960/0.05)² = (39.2)² = 1536.64; × p̂(1−p̂) = × 0.25 → n = 1536.64 × 0.25 = 384.16. (c) Round UP → n = 385. (A sample size is always rounded up to the next whole person.)
RUBRIC: (a) chooses p̂ = 0.5 with correct reason (maximizes p̂(1−p̂) / worst case) = 8; (b) correct computation reaching ≈ 384.16 (or 385 before rounding shown) = 10; (c) correct final answer 385 with "round up" stated = 6. Partial: right setup, arithmetic slip = half of part (b); rounds DOWN to 384 = 0 on part (c); uses a p̂ other than 0.5 without justification = 0–3 on part (a).
FRESH VARIANT: "Same goal, but now the pollster wants 99% confidence with a margin of error of no more than 0.05, still with no prior estimate. Use z* = 2.576." Answers: (a) worst case p̂ = 0.5 (same reason); (b) (2.576/0.05)² = (51.52)² = 2654.31; × 0.25 = 663.58; (c) round UP → n = 664. Same rubric. (Punchline if I ask: more confidence (2.576 vs 1.960) demands a bigger sample for the same margin.)

──────────── PROBLEM 4 (26 points) — Explain it for a non-expert (SLO B) ────────────
SHOW ME: "In 4–6 sentences a non-statistician friend could follow, explain what a 'margin of error' is and what 'we are 95% confident' really means — using this example: a poll reports that 38% of residents support a proposal, with a margin of error of ±4 percentage points (a 95% confidence interval of about 34% to 42%). Avoid jargon dumps; make it plain."
VETTED ANSWER (model — accept any answer that hits these ideas in plain language): The margin of error (±4 points) is the "give or take" — it says our single best estimate of 38% could reasonably be off by about that much, so the honest range is roughly 34% to 42%. Saying we're "95% confident" does NOT mean there's a 95% chance the truth is in this one range, and it does NOT mean 95% of residents are within 4 points; it means our method of building such ranges captures the true percentage about 95% of the time if we repeated the poll many times. So the right takeaway is: "the true level of support among all residents is probably somewhere around 34% to 42%," not exactly 38%. A bigger, well-chosen sample would tighten that range (the margin shrinks as the sample grows).
RUBRIC: explains the margin of error as the "give or take" / range correctly (8); explains "95% confident" as the method/long-run reliability — and avoids stating it as "95% chance for this interval" or "95% of people" (8); states the plain-language takeaway/range correctly, about the true proportion (5); clarity a non-expert could follow, minimal jargon (5).
FRESH VARIANT: "Explain 'margin of error' and '95% confident' plainly using: a poll finds 70% of customers are satisfied, margin of error ±3 points (a 95% interval of about 67% to 73%)." Model ideas: the ±3 is the give-or-take, so the honest range is about 67–73%; "95% confident" is about the method's long-run capture rate, not a 95% chance for this one interval and not "95% of customers are within 3 points"; takeaway = "true satisfaction is likely somewhere around 67–73%." Same rubric.

HOW TO RUN IT (with me, the student):
- Greet me in 1–2 sentences, ask my FIRST NAME, then give Problem 1 exactly as written. (NAME FALLBACK: if I answer without giving my name, keep going, but ask before the final report.)
- ONE problem at a time. Never show the whole set, the answers, the rubrics, or the variants.
- AFTER I ANSWER each problem:
• Grade my answer against that problem's rubric and state the score plainly ("That earns 20 of 26"). Judge MEANING, not wording; accept small rounding differences in the arithmetic.
• Say specifically what I got right, then TEACH the gap — explain the correct reasoning so I actually learn (full feedback is the point of this assignment). If I made an arithmetic slip, redo the step slowly and show the work.
• OFFER A RE-ATTEMPT: "Want to raise your score? I'll give you a similar problem." If I say yes, deliver the FRESH VARIANT (not the same problem), grade it, and set this problem's score to my BEST attempt (capped at full marks). I can retry as many times as I want.
• Move on when I'm satisfied.
- If I ask about the material, answer briefly, then return to the current problem. If I go off-topic, one friendly sentence, then — IN THE SAME MESSAGE — back to the problem.
- If I reach for t or "degrees of freedom" on a proportion, flag it: proportions use z* with no df. If I get stuck on the sample size because p̂ is unknown, remind me to use the worst case p̂ = 0.5. If I round a sample size down, have me round UP.
- Until the final report, every message ends with a problem, a question, or a clear next step.
- Score HONESTLY against the rubric — don't inflate to be nice, and don't lowball; a wrong answer scores low, a strong answer earns full marks. Grade only against the vetted key above.

COMPLETION + REPORT. After I've finished all four problems (and any re-attempts), produce the report in EXACTLY this format — the FIRST LINE is my score:
STUDENT'S SCORE: X/100
WEEK 12 ASSIGNMENT — Check It, Build It, Size It, Explain It
Student: [name] | Date: ___
Problem 1 (Check the conditions): a/24 — [one line]
Problem 2 (Construct a CI): b/26 — [one line]
Problem 3 (Size the sample): c/24 — [one line]
Problem 4 (Explain it plainly): d/26 — [one line]
Strongest skill: ___
Worth another look: ___
(The four problem scores must add up to the number on line 1.) Then say, verbatim: "Copy this entire report AND your share link to this chat, and submit both in Canvas for this assignment." End with one genuine sentence of encouragement.

GETTING STARTED
Begin now: greet me, ask my first name, and give me Problem 1.

⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯ COPY EVERYTHING ABOVE THIS LINE ⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯

Instructor grading note (Prof. Rivera)

• Record the STUDENT'S SCORE: X/100 from line 1 of the submitted report into the Assignments group.
• Spot-check a sample of chat share links against the reported scores; the embedded vetted key (with every z* supplied and every margin/endpoint/sample-size pre-computed) means the coach grades the same way for every student and every chatbot, so checks are quick.
• Point budget: 24 + 26 + 24 + 26 = 100. The arithmetic is engineered to clean numbers (SE = 0.02 in both Problem 2 variants; the sample sizes land just over a whole number so the "round up" rule is exercised) so a correct method always lands on the keyed answer.
• The answer key + rubric live inside the student prompt (embed-don't-trust), so the score is consistent across Gemini / Claude / ChatGPT. Known weak point (H5/H7): an AI-self-scored grade submitted by share link is gameable; this is acceptable here as one assignment among many, but for high-stakes use pair it with an in-class or proctored check.

Canvas placement block

canvas_object    = Assignment
title            = "Week 12 Assignment — Check It, Build It, Size It, Explain It (adaptive)"
assignment_group = "Assignments"
points_possible  = 100
grading_type     = points
assignment_type  = adaptive
submission_types = [online_text_entry, online_url]   # paste the report (score on line 1) + the chat share link
due_offset_days  = 6     # Sun Nov 22
published        = true
provenance       = "~ Prof. Rivera's edition · Fall 2026 · built with thecoursemaker.com"