Week 12 — Quiz (auto-graded) · Public Opinion, Political Behavior & the Media

Course: Introduction to Political Science (POLS 1) · Silver Oak University (fictional sample) · Prof. Halloran
Objectives tested: Objective 6 — public opinion; random sampling; margin of error; MoE vs. bias; political socialization; turnout; media effects.
Points: 10 (1 each) · Assignment group: Quizzes (10% of grade) · Due: end of Module 12.

This is the human-readable quiz with its vetted answer key and feedback. The import-ready Classic QTI is in F-quiz-week-12-qti.xml (generated by the shared validated script — parses with 10 items, every single-answer item exactly one correct). Fact-and-source-accuracy gate — PASS: the MoE formula and all three worked values (Q1's n = 1,600 computation re-run in Python; the FACTS_PACK's n = 1,000/400/2,500 reference values), the delegate/trustee definitions, the political-socialization agents, and the agenda-setting/framing definitions were each verified against the record. The Canvas placement block is at the bottom of this file.

Blueprint

No trick questions; distractors target the Week 12 misconceptions named in the lecture outline (MoE vs. bias; MoE as a hard guarantee; self-selected online polls treated as trustworthy; agenda-setting confused with "the media dictates conclusions"; using the textbook formula instead of a poll's own stated MoE).

Questions, key, and feedback

Q1 (MC). A national poll surveys a random sample of 1,600 U.S. adults. Using MoE = 1.96 × √(0.25/n) at 95% confidence, the margin of error is closest to:
- A. ±1.0 percentage points
- B. ±2.5 percentage points ✅
- C. ±4.0 percentage points
- D. ±6.0 percentage points
Feedback: 1.96 × √(0.25/1,600) = 1.96 × √0.00015625 = 1.96 × 0.0125 = 0.0245 → ≈ ±2.5 points (re-verified in Python). Notice the pattern from class: 1,600 = 4 × 400, and 400's MoE (±4.9) is almost exactly double 1,600's (±2.5) — quadrupling the sample halves the error.

Q2 (True/False). "If Poll A has a much larger sample size than Poll B, Poll A is automatically more trustworthy — regardless of how either poll's respondents were selected."
- True
- False ✅
Feedback: False. Sample size alone only controls margin of error (precision). If Poll A's larger sample was self-selected (not randomly drawn), it can be more biased, not more trustworthy, than Poll B's smaller random sample. Bigger fixes precision, not bias.

Q3 (MC). Before citing any poll number, a political scientist checks four facts. Which of the following is NOT one of them?
- A. Who was sampled, and how many (the sample and its size)
- B. Exactly when the poll was in the field
- C. Which news outlet first reported the story ✅
- D. The poll's stated margin of error
Feedback: The four load-bearing facts are the sample, the field dates, how it was sampled (random? from what frame?), and the margin of error. Which outlet happened to cover the story first says nothing about the poll's own reliability.

Q4 (MC). Random sampling lets a poll of a couple thousand people estimate the opinion of hundreds of millions primarily because:
- A. Pollsters interview people from every state to be fair
- B. Every person in the population has a known, nonzero chance of being selected, making the sample's answers an unbiased estimate of the population's ✅
- C. Larger populations are naturally easier to predict than smaller ones
- D. Modern computers can process any sample size instantly
- Feedback: The mathematical guarantee behind polling rests on known selection probability, not on interviewing "enough" people in an intuitive sense or on computing power. That's what separates a real random-sample poll from a self-selected online poll of any size.

Q5 (Matching). Match each term to its correct description.
| Term | Description |
|---|---|
| Margin of error | How much a poll's reported number could differ from the true population value, at a stated confidence level |
| Bias (in polling) | Systematic error from a flawed method — e.g., a non-random sample — that a larger sample size cannot fix |
| Political socialization | The lifelong process by which people acquire their political attitudes and identities |
| Agenda-setting | Media influence over which issues the public sees as important, not over what conclusions to reach |
| Delegate (representation model) | The view that an elected official should closely follow constituents' current expressed wishes |
Feedback: Keep these five straight by function: MoE is about precision; bias is about method; socialization is about origin (where views come from); agenda-setting is about salience (what feels important); delegate is one of two classic representation models (the other, trustee, follows independent judgment).

Q6 (MC). Which agent of political socialization is generally the strongest early influence on a person's party identification?
- A. A person's employer
- B. Family ✅
- C. A single news broadcast
- D. A ballot initiative
Feedback: Family is present earliest and most consistently in a person's life, and research consistently finds it the strongest early influence on basic political orientation — though its grip loosens somewhat with independent life experience.

Q7 (MC). Compared to midterm-election turnout, U.S. presidential-election turnout is:
- A. Reliably higher ✅
- B. Reliably lower
- C. Essentially identical
- D. Unpredictable — no documented pattern exists
Feedback: This is a well-documented, non-partisan empirical pattern: presidential-year elections consistently draw higher turnout than midterm elections, across many election cycles.

Q8 (Multiple answer — select all that apply). Which of the following are accurate statements about agenda-setting research, as presented in class?
- A. It finds the media has measurable influence over which issues the public sees as important ✅
- B. It proves the media can dictate exactly what conclusions people reach on an issue
- C. Critics note the effect is limited by audiences actively selecting and resisting sources ✅
- D. Its reach may be fragmented today by many competing outlets and algorithmic feeds ✅
- E. It has been fully disproven by modern research
Feedback: Agenda-setting is a real, well-established finding (A) about salience (what feels important), not conclusions (so B is false) — and it comes with real, documented limits (C, D), not a full disproof (E is false). Present the finding with its critics, always.

Q9 (MC). In the classic delegate vs. trustee debate about representation, a trustee-style representative believes that an elected official should:
- A. Vote however the most recent poll of their district indicates
- B. Exercise independent judgment on constituents' behalf, even against current opinion, because they were elected partly for that judgment ✅
- C. Never consult public opinion under any circumstances
- D. Resign if their judgment ever conflicts with a majority poll result
Feedback: The trustee model (contrasted with the delegate model, which does closely follow current opinion) holds that representatives owe constituents their considered judgment, not a running tally of the latest poll — a genuinely contested, evenhandedly-presented question, not a settled one.

Q10 (True/False). "The exact wording and order of questions in a poll can measurably shift the results, which is why reputable pollsters publish their exact question wording."
- True ✅
- False
Feedback: True. Question-wording and order effects are well-documented (e.g., "government assistance" vs. "welfare" can shift results by meaningful margins) — exactly why checking a pollster's published exact wording is part of reading a release responsibly.

Answer key (quick reference)

Quality gate (self-checked): each single-answer item has exactly one correct option; the multiple-answer item lists the three true statements (A, C, D) and requires B and E to be left unselected; the matching item pairs each of the five terms with its real function. Fact-and-source-accuracy gate — PASS: the MoE formula and Q1's n = 1,600 arithmetic were re-run in Python (1.96 × √(0.25/1600) = 0.0245 → ±2.45, rounds to ±2.5); the presidential-vs-midterm turnout pattern, the family-as-strongest-early-socialization-agent finding, and the agenda-setting/framing definitions were each checked against the record; no invented statistic appears. Evenhandedness check: no item asks which representation model or media-effects position is "correct"; Q8 and Q9 test what the research/positions claim, not which is right.

Item-bank entries (for variants + the final)

All ten items are tagged course=POLS1 · week=12 · objective=6 · topic=public-opinion-polling-media and deposited in Item Bank: Week 12 — Public Opinion, Political Behavior & the Media. The final (Week 16) and any per-term variant updates draw fresh items from this bank. (Tags: q1 moe-computation, q2 bias-vs-size, q3 four-facts, q4 random-sampling-why, q5 term-matching, q6 socialization-family, q7 turnout-presidential-midterm, q8 agenda-setting-multi, q9 delegate-trustee, q10 wording-order-effects.)

Canvas placement block

canvas_object    = Quizzes::Quiz
title            = "Week 12 Quiz — Public Opinion, Political Behavior & the Media"
assignment_group = "Quizzes"
points_possible  = 10
grading_type     = points
due_offset_days  = 6        # 6 days after module start
published        = true
shuffle_answers  = true
provenance       = "~ Prof. Halloran's edition · Fall 2026 · built with thecoursemaker.com"