College Board CED p.239 원문 + AP Stat 채점기준에서 점수 받는 답안 vs 감점 답안 예시
College Board AP Statistics CED p.20 / p.238 표기.
| Skill 1: Selecting Statistical Methods | 15–23% |
| Skill 2: Data Analysis | 15–23% |
| Skill 3: Using Probability and Simulation | 30–40% |
| Skill 4: Statistical Argumentation | 25–35% |
| U1 Exploring One-Variable Data | 15–23% |
| U2 Exploring Two-Variable Data | 5–7% |
| U3 Collecting Data | 12–15% |
| U4 Probability·Random Var.·Distrib. | 10–20% |
| U5 Sampling Distributions | 7–12% |
| U6 Inference: Proportions | 12–15% |
| U7 Inference: Means | 10–18% |
| U8 Inference: Chi-Square | 2–5% |
| U9 Inference: Slopes | 2–5% |
문제: "Identify the explanatory variable in this study."
답: "The amount of caffeine (in mg) consumed by each subject."
→ 단답으로 끝. 1점.
"이 연구에서 설명변수는 카페인 양인데, 왜냐하면 연구자가 이 변수를 조작했고, 결과변수인 반응시간에 영향을 주는지 보려 했기 때문에... caffeine amount가 explanatory."
→ 정답이지만 시간 4배 소비. Q1당 13분 룰 위반.
H₀: μ = 65 H_a: μ > 65
where μ = the true mean
resting heart rate (in bpm) of
all HAFS students.
→ 양쪽 가설 + 모수 + 단위 + 모집단. 1점.
H₀: x̄ = 65
H_a: x̄ > 65
→ Statistic을 가설에 사용 → 자동 0점. x̄는 표본통계량.
H₀: μ = 65, H_a: μ ≠ 65
→ 모수 정의 (in context) 누락 → -1점.
"The distribution of resting heart rates is roughly symmetric and unimodal. There are no apparent outliers. The mean resting heart rate is approximately 72 bpm, and the standard deviation is approximately 8 bpm."
→ Shape ✓ Outliers ✓ Center ✓ Spread ✓ Context ✓. 만점.
"The distribution is symmetric with mean 72."
→ Outlier 언급 없음, Spread 없음, "of what?" 맥락 없음. 4점 중 1~2점만.
"It is skewed right and the average is 72 with SD 8."
→ 오른쪽 치우침인데 평균·SD 사용 → 자동 -1점 (median/IQR 써야).
"The distribution of method A is more right-skewed than method B, which is roughly symmetric. The median of A (12) is higher than the median of B (8). The IQR of A (6) is larger than the IQR of B (3), indicating method A has greater variability."
→ Shape, Center, Spread 모두 비교어 사용. 만점.
"Method A is right-skewed with median 12 and IQR 6. Method B is symmetric with median 8 and IQR 3."
→ 정보는 있지만 비교어가 하나도 없음. AP rubric: "Comparing requires explicit comparative language" → 0점.
문제: "Calculate the test statistic and the P-value."
t = (x̄ − μ₀) / (s/√n)
= (74.2 − 72) / (8/√40)
= 2.2 / 1.265
= 1.74
P-value = P(t > 1.74), df = 39
= 0.0449
(using tcdf(1.74, 99, 39))
→ 공식 + 대입 + 계산 + 결과 + 계산기 함수 명시. 2점.
t = 1.74
P-value = 0.0449
→ 답이 정답이라도 대입 과정 없음 → 0점 또는 부분점만. CED 원문: "with properly substituted numbers" 위반.
"Yes, the data support the claim that the mean resting heart rate is greater than 72 bpm. Since P-value = 0.045 < α = 0.05, we reject H₀ and conclude there is convincing evidence at the 0.05 level."
→ 결론(yes) + 정량 근거(P-value vs α) + 맥락. 만점.
"Yes, the data support the claim."
→ 근거 없음. -1~-2점.
"P-value = 0.045."
→ 근거만 있고 결론(yes/no) 없음. -1점.
"We are __% confident that the interval from __ to __
captures the true [parameter, in context, with units]."
예: "We are 95% confident that the interval from 70.1 to 76.3 bpm
captures the true mean resting heart rate of all HAFS students."
② 신뢰수준 해석 (Interpret the confidence level)
"If we were to take many random samples of the same size and
construct a __% confidence interval from each, about __% of those
intervals would capture the true [parameter, in context]."
③ P-value 해석 (Interpret the P-value)
"Assuming H₀ is true (i.e., [μ = ___, in context]), the probability
of observing a sample [statistic] as extreme as or more extreme
than the one observed is [P-value]."
④ Slope 해석 (Interpret slope b)
"For each additional [1 unit of x, in context], the predicted
[y, in context] increases/decreases by [b] [units of y]."
"We are 95% confident that the interval from 70.1 to 76.3 bpm captures the true mean resting heart rate of all HAFS students."
→ 신뢰수준 + 구간 + 단위 + 모수 + 모집단(in context). 만점.
"There is a 95% probability that μ is between 70.1 and 76.3."
→ "probability that μ" = 자동 0점. μ는 고정값이지 확률변수가 아님. "capture"라는 단어 필수.
"95% of HAFS students have heart rates between 70.1 and 76.3."
→ 모수(평균)와 개체값을 혼동 → 자동 0점.
"Random assignment of subjects to groups is important because it tends to balance both known and unknown confounding variables across the treatment and control groups, so any observed difference in mean outcomes can be attributed to the treatment rather than to lurking variables."
→ "because + so" 인과 사슬 + 통계적 원리. 만점.
"Random assignment makes the experiment fair."
→ "fair"라는 표현은 통계 용어 아님. 메커니즘(confounding 균형) 없음. 0점.
"The t-procedure is appropriate. (1) The data come from a random sample, satisfying the independence/randomness condition. (2) The sample size is n=40 ≥ 30, so by the Central Limit Theorem, the sampling distribution of x̄ is approximately normal. (3) The sample is <10% of the population of all HAFS students, satisfying the 10% condition for independence."
→ Claim + 3개 조건 각각에 대한 evidence + statistical reasoning. 만점.
"It's appropriate because we have enough data."
→ 통계적 논리 없음. 조건 명시 X. 0점.
"Random + n≥30 + independent."
→ 키워드만 나열, "어떻게 충족되는지(how)"의 evidence 없음 → 부분점만.
| Task Verb | 한국어 | 핵심 행동 | 0점되는 실수 |
|---|---|---|---|
| Identify / Indicate / Circle | 식별·표시 | 단답. 설명 X. | (없음 — 길게 써도 점수) |
| State | 제시 (가설·결론) | H₀, H_a 둘 다 + parameter in context | x̄, p̂ 같은 statistic 사용 → 0점 |
| Describe | 기술 (분포·산점도) | SOCS 4요소 (또는 DUFS) + context | 요소 누락, 비교어 X, "of what?" 없음 |
| Compare | 비교 | 비교어("higher than" 등) + 양쪽 SOCS | 비교어 0개 = 0점 (단순 나열) |
| Calculate | 계산 | 공식 + 대입 + 답 + 단위 | work 없으면 0점 |
| Construct / Complete | 그래프·표 작성 | 축·라벨·단위·척도·outlier | 축 라벨 누락, 척도 불균등 |
| Determine | 판단·결정 | 결론 + 수치 근거 (P-value vs α) | 결론만 또는 근거만 |
| Interpret | 해석 | 정해진 템플릿 + in context + units | "probability that μ" = 자동 0점 |
| Explain | 설명 | "because + so" 인과 사슬 | 단답 시 0점 |
| Justify | 정당화 | Claim + Evidence + Stat. Reasoning | 키워드만 나열, "convincing" 누락 |
| # | 실수 | 잘못된 예 | 옳은 형태 |
|---|---|---|---|
| 1 | Statistic을 가설에 사용 | H₀: x̄ = 65 | H₀: μ = 65 (parameter) |
| 2 | "probability that μ" 사용 | "95% probability μ is in (70, 76)" | "95% confident the interval captures μ" |
| 3 | P-value 해석 오류 | "P-value = probability H₀ is true" | "Assuming H₀ true, prob. of data as extreme..." |
| 4 | 비교어 누락 | "A: median 12, B: median 8" | "A의 median은 B보다 높다 (12 vs 8)" |
| 5 | SOCS 요소 누락 | "Symmetric, mean=72" | Shape + Outliers + Center + Spread + context |
| 6 | 치우친 분포에 mean·SD 사용 | "Right-skewed, mean=12, SD=4" | "Right-skewed, median=10, IQR=6" |
| 7 | "convincing evidence" 누락 | "There is strong/good evidence..." | "There is convincing evidence..." |
| 순서 | 자료 | 용도 |
|---|---|---|
| 1 | 📝 Task Verbs (지금 이 자료) | FRQ 답안 형태 — 30분 정독 |
| 2 | 📋 AP Stat 치트시트 | 공식·정의·조건 1쪽 압축본 — 20분 |
| 3 | 📊 FRQ 채점 가이드 | 실제 rubric 분석 — 30분 |
| 4 | ✍️ FRQ 작문 드릴 | Interpret·Justify 템플릿 손에 익히기 — 40분 |
| 5 | 🧪 Q6 Investigative Task 트레이너 | 가장 어려운 Q6 (25분) 대비 — 60분 |
| 6 | 🎓 AP Stat Mastery | 전 단원 통합 학습툴 (있으면) |