Common Mistakes in A-Level Statistics and How to Correct Them | A-Level AQA 统计：常见误区与纠正方法

📚 Common Mistakes in A-Level Statistics and How to Correct Them | A-Level AQA 统计：常见误区与纠正方法

Many A-Level Statistics students repeatedly stumble over the same conceptual traps, costing them valuable marks on the AQA exam. From muddling sample and population variance to misreading the p‑value, these errors often stem from half‑remembered formulas or rushed interpretations. This article walks through the most frequent pitfalls and shows you exactly how to set them right, so your reasoning becomes sharp, precise, and examiner‑friendly.

许多 A-Level 统计学生反复掉进相同的概念陷阱，导致在 AQA 考试中丢失宝贵分数。从混淆样本方差与总体方差，到误解 p 值，这些错误往往源于对公式的模糊记忆或仓促的解读。本文将梳理最常见的误区，并给出准确的纠正方法，让你的推理变得敏锐、精确，并符合考官的要求。

1. Confusing Sample Variance with Population Variance | 混淆样本方差与总体方差

One of the earliest sticking points is when to divide by n and when to divide by (n – 1). If you are working with a population or a theoretical distribution, the variance σ² = Σ(x – μ)² ÷ n. However, for a sample, the unbiased estimator of the population variance is s² = Σ(x – x̄)² ÷ (n – 1). Examiners frequently test this distinction, especially when the question moves from descriptive statistics to hypothesis testing.

最常见的绊脚石之一是什么时候除以 n，什么时候除以 (n – 1)。如果你处理的是总体或理论分布，方差为 σ² = Σ(x – μ)² ÷ n。但是，对于样本，总体方差的无偏估计量是 s² = Σ(x – x̄)² ÷ (n – 1)。考官经常测试这一区别，尤其是当问题从描述性统计过渡到假设检验时。

In AQA mechanics, forgetting to use n – 1 for the sample variance will lead to a slightly smaller spread, which in turn can inflate your test statistic and distort the p‑value. Always check: if the data are a sample drawn from a larger population and you are asked to estimate the population variance, the (n – 1) divisor is mandatory.

在 AQA 的题目中，忘记对样本方差使用 n – 1 会导致离散程度被略微低估，进而夸大检验统计量并扭曲 p 值。务必检查：如果数据是从较大总体中抽取的样本，并要求你估计总体方差，则必须采用 (n – 1) 作为除数。

2. Misinterpreting the Significance Level | 误解显著性水平

Many students treat the significance level α as the probability that the null hypothesis is true. This is a serious error. The significance level is actually the probability of rejecting H₀ when H₀ is true – a Type I error. It does not tell you how likely H₀ is to be correct; it only quantifies the risk you are willing to take of a false rejection.

许多学生将显著性水平 α 视为原假设为真的概率。这是一个严重错误。显著性水平其实是在 H₀ 为真的条件下拒绝 H₀ 的概率——即第一类错误。它并不告诉你 H₀ 正确的可能性；它只量化了你愿意承担的误拒风险。

In AQA exam questions, you might be asked to state what a 5% significance level means. The correct answer: there is a 5% chance of concluding that the null hypothesis is false when it is actually true. Avoid wording that suggests α refers to the probability of H₀.

在 AQA 考试题中，你可能会被问到 5% 显著性水平的含义。正确的答案是：在原假设实际为真时，有 5% 的概率得出结论认为它是错误的。避免使用任何暗示 α 是 H₀ 概率的措辞。

3. Treating ‘Fail to Reject’ as ‘Accept H₀’ | 把“不拒绝”当成“接受 H₀”

Hypothesis testing never proves a null hypothesis; it only assesses whether the evidence is strong enough to reject it. When the p‑value exceeds α, we say we do not reject H₀, not that we accept it. The phrase “accept H₀” implies proof of no effect, which the test cannot provide.

假设检验永远无法证明原假设成立；它只能评估证据是否强到足以拒绝它。当 p 值大于 α 时，我们说未拒绝 H₀，而不是接受它。“接受 H₀” 这个说法暗示着证明了没有效应，而检验无法提供这样的证明。

On AQA papers, writing “accept H₀” when the result is not significant can be marked down. The safer phrasing is “there is insufficient evidence to reject H₀” or “the result is not significant at the α level.” This subtlety matters because it shows you understand the limits of statistical inference.

在 AQA 试卷上，当结果不显著时写“接受 H₀” 可能会被扣分。更稳妥的措辞是“没有充分证据拒绝 H₀” 或“结果在 α 水平下不显著”。这种细微差别很重要，因为它表明你理解统计推断的局限性。

4. Getting the Direction of Hypotheses Wrong | 设错假设的方向

A common error is to confuse the null and alternative hypotheses, especially in one‑tailed tests. H₀ always contains an equality (=, ≤, or ≥). H₁ is what you are trying to prove: a strict inequality (<, >, or ≠). If the question asks “has the mean increased?”, H₁: μ > μ₀ and H₀: μ ≤ μ₀. Writing H₁: μ ≥ μ₀ would lose marks because the alternative must reflect the research claim without equality.

一个常见错误是混淆原假设和备择假设，尤其是在单尾检验中。H₀ 总是包含等号（=、≤ 或 ≥）。H₁ 是你试图证明的：严格的不等式（<、> 或 ≠）。如果问题问“均值是否增加了？”，那么 H₁: μ > μ₀，H₀: μ ≤ μ₀。写成 H₁: μ ≥ μ₀ 将会失分，因为备择假设必须反映研究主张，不含等号。

Students sometimes flip the direction when reading a scenario. Always identify the claim as H₁ and then set H₀ as its complement. This prevents mistakes in the rejection region and p‑value calculation.

学生有时在阅读情景时会把方向搞反。始终将主张识别为 H₁，然后将 H₀ 设为其补集。这能防止在拒绝域和 p 值计算中出现错误。

5. Misunderstanding the p‑value | 误解 p 值

The p‑value is not the probability that the null hypothesis is true, nor is it the probability that the results happened by chance alone, although it is often loosely described that way. Formally, the p‑value is the probability of obtaining a test statistic at least as extreme as the one observed, assuming H₀ is true. The phrase “assuming H₀ is true” is crucial and should appear in your definition.

p 值不是原假设为真的概率，也不是结果仅仅由偶然性造成的概率——尽管人们经常这样粗略地描述。形式上，p 值是在假定 H₀ 为真的条件下，获得至少与实际观测值一样极端的检验统计量的概率。“假定 H₀ 为真” 这一短语至关重要，应该出现在你的定义中。

If a student writes “p = 0.03 means there is a 3% chance the null is true,” that is conceptually wrong and will lose marks. Practice writing: “If H₀ were true, the probability of seeing a result this extreme (or more) would be 0.03.”

如果学生写“p = 0.03 意味着原假设有 3% 的概率为真”，这在概念上是错误的，并且会失分。请练习写作：“如果 H₀ 为真，观察到如此极端（或更极端）结果的概率为 0.03。”

6. Confusing Correlation with Causation | 混淆相关与因果

After you calculate a PMCC (r) and find it is significant, it is tempting to say the variable X causes the change in Y. But correlation measures linear association only. A strong r might be due to a lurking variable, chance, or reverse causation. In AQA questions that ask for interpretation, you should restrict your comments to the strength and direction of the linear relationship, not causation.

当计算 PMCC (r) 并发现它显著后，人们很容易说变量 X 导致了 Y 的变化。但相关只衡量线性关联。一个

Published by TutorHao | A-Level 统计 Revision Series | aleveler.com

更多咨询请联系16621398022（同微信）