📚 Common Misconceptions and Corrections in AS WJEC Statistics | AS WJEC 统计：常见误区与纠正方法
AS WJEC Statistics covers a wide range of topics from data presentation to probability distributions. Many students lose marks not because they lack understanding, but because they fall into common traps that can be avoided with clear strategies. This article identifies the most frequent misconceptions and provides step-by-step corrections to help you score higher in your exam.
AS WJEC 统计学涵盖从数据展示到概率分布的广泛主题。许多学生丢分并非因为不理解，而是陷入了可以避免的常见陷阱。本文指出最常见的误区并逐步纠正，帮助你在考试中取得更高分数。
1. Stem-and-Leaf Diagram Key Errors | 茎叶图键值错误
A stem-and-leaf diagram without a key is meaningless to the examiner. Students often draw the diagram correctly but either omit the key or write something like ‘6 | 2 = 6.2’ without specifying units, which creates ambiguity. In back-to-back stem-and-leaf diagrams, forgetting to add titles for each side can lead to confusion about which data set is which.
没有键值的茎叶图对考官来说毫无意义。学生常常画对了图，却要么遗漏键值，要么写成“6|2 = 6.2”却不指定单位，造成歧义。在背靠背茎叶图中，忘记给两侧添加标题可能导致无法区分数据集。
Correction: Always write a key such as ‘6 | 2 represents 62 marks’ near the diagram. For back-to-back, label the left leaves as one group, the stem in the centre, and the right leaves as the other group. For example, ‘Leaf (Girls) | Stem | Leaf (Boys)’. The key should explain both sides if the scaling differs.
纠正： 始终在图的旁边写出键值，如“6 | 2 表示 62 分”。背靠背图要分别标注左侧叶子为组一，中间为茎，右侧叶子为组二，例如“叶子(女生) | 茎 | 叶子(男生)”。若两侧尺度不同，需分别说明键值。
2. Misusing Mean, Median and Mode in Skewed Data | 偏态数据中误用均值、中位数和众数
A frequent error is quoting the mean as the best measure of location for heavily skewed data, such as income or house prices. In positively skewed distributions, the mean is pulled towards the tail and overestimates the typical value. Students also incorrectly state the order of the three averages without sketching the distribution.
一个常见错误是对于严重偏态的数据（如收入或房价）仍引用均值作为最佳位置度量。在正偏态分布中，均值被拉向尾端，高估了典型值。学生还会错误地陈述三个平均数的顺序而不结合分布图形。
Correction: For skewed data, use the median as the preferred measure because it is resistant to outliers. The mode lies at the peak, the median divides the area in half, and the mean is furthest towards the skew. In a positive skew: mode < median < mean. Always relate the statement to the shape of the data.
纠正： 对于偏态数据，使用中位数作为首选度量，因为它不受异常值影响。众数位于峰值处，中位数将面积分为两半，均值最偏向偏斜方向。正偏态时：众数 < 中位数 < 均值。始终结合数据形状来陈述。
3. Variance: Dividing by n or n–1? | 方差：除以 n 还是 n–1？
WJEC questions often provide data and ask for the variance or standard deviation. A major pitfall is automatically using the formula with n in the denominator for all cases. Students mix up the population variance σ² = Σ(x – μ)²/n with the sample variance s² = Σ(x – x̄)²/(n – 1). When the data is a sample intended to estimate a population, using n instead of n–1 underestimates the variability.
WJEC 试题经常给出数据并要求计算方差或标准差。一个主要陷阱是不分情况一律使用分母为 n 的公式。学生混淆了总体方差 σ² = Σ(x – μ)²/n 与样本方差 s² = Σ(x – x̄)²/(n – 1)。当数据是用于估计总体的样本时，用 n 而不是 n–1 会低估变异性。
Correction: Read the question carefully. If the data represents the entire population (e.g. all customers of a small store on a particular day), use n. If the data is a sample (e.g. ‘a random sample of 50 students’), use n–1. Many formulas on the WJEC formula sheet explicitly show the divisor as n or n–1, so check which one is required. For frequency distributions, the same rule applies: divide by Σf or Σf – 1 accordingly.
纠正： 仔细读题。如果数据代表整个总体（例如某小店某天的所有顾客），用 n。如果数据是样本（例如“随机抽取的50名学生”），用 n–1。WJEC 公式表明确显示了分母为 n 或 n–1，请确认题目要求。对于频数分布，规则相同：相应地除以 Σf 或 Σf – 1。
4. Probability Pitfalls: Mutually Exclusive vs Independent | 概率陷阱：互斥与独立
Students often think mutually exclusive events are automatically independent, or vice versa. They misuse the addition rule P(A ∪ B) = P(A) + P(B) – P(A ∩ B) by forgetting the subtraction when events are not mutually exclusive. Another error is assuming P(A ∩ B) = P(A) × P(B) without checking for independence, or incorrectly using this product rule for mutually exclusive events where P(A ∩ B) = 0.
学生常认为互斥事件自然独立，或反之。他们使用加法公式 P(A ∪ B) = P(A) + P(B) – P(A ∩ B) 时，因事件不互斥却忘记减去交集部分。另一个错误是在未验证独立性的情况下假设 P(A ∩ B) = P(A) × P(B)，或者对互斥事件错误地使用该乘积规则，而互斥事件的 P(A ∩ B) = 0。
Correction: Mutually exclusive means A and B cannot occur together: P(A ∩ B) = 0. Independent means the occurrence of A does not affect the probability of B: P(B|A) = P(B). Mutually exclusive events with non-zero probabilities are never independent, because if A occurs, B cannot occur. Always use the general addition rule unless you are certain events are mutually exclusive. For intersection, use P(A) × P(B) only if independence is stated or proven.
纠正： 互斥意味着 A 和 B 不能同时发生：P(A ∩ B) = 0。独立意味着 A 的发生不影响 B 的概率：P(B|A) = P(B)。具有非零概率的互斥事件永不独立，因为如果 A 发生，B 就不能发生。除非确定事件互斥，否则始终使用通用加法法则。对于交集，仅在题目申明或验证了独立性时才使用 P(A) × P(B)。
5. Tree Diagrams and Conditional Probability | 树图与条件概率
When constructing tree diagrams, a common slip is writing the wrong conditional probabilities on the second branches. Students may use simple probabilities instead of conditional ones, e.g. P(B|A) rather than just P(B). Additionally, when finding the conditional probability of an event given another event, many forget to divide by the probability of the conditioning event, applying P(A ∩ B) as the final answer.
构建树图时，一个常见疏忽是在第二级分支上写错条件概率。学生可能使用简单概率而非条件概率，例如写 P(B) 而不是 P(B|A)。此外，在求一个事件在另一事件发生条件下的条件概率时，许多人忘记除以条件事件的概率，直接将 P(A ∩ B) 作为最终答案。
Correction: Label the first set of branches with P(A) and P(A’). On the next level, every branch must be a conditional probability such as P(B|A). To calculate P(A ∩ B), multiply along the required branch path. To find a conditional probability P(A|B), use the formula P(A ∩ B)/P(B), where P(B) is found by summing the probabilities of all paths leading to B. Always check that the sum of probabilities on corresponding branches equals 1.
纠正： 第一级分支标注 P(A) 和 P(A’)。往下的每一级分支必须都是条件概率，例如 P(B|A)。计算 P(A ∩ B) 时，沿所需路径相乘。求条件概率 P(A|B) 时，使用公式 P(A ∩ B)/P(B)，其中 P(B) 通过将所有通往 B 的路径概率相加得到。始终检查同级分支概率之和为 1。
6. Linear Transformations of Random Variables | 随机变量的线性变换
A classic mistake is adding a constant to the variance. For a discrete random variable X, if Y = aX + b, then Var(Y) = a²Var(X), but students often write Var(Y) = a²Var(X) + b or even aVar(X) + b. Similarly, with the expected value E(Y) = aE(X) + b, they may forget to add b. In problems involving E(X²) from a coded variable, the uncoding step is frequently mishandled.
一个经典错误是把常数加到方差上。对于离散随机变量 X，若 Y = aX + b，则 Var(Y) = a²Var(X)，但学生常写成 Var(Y) = a²Var(X) + b 甚至 aVar(X) + b。类似地，期望值 E(Y) = aE(X) + b 时，可能忘记加 b。在涉及从编码变量求 E(X²) 的问题中，解码步骤经常处理不当。
Correction: Remember that adding a constant shifts the mean but has no effect on spread, so b disappears in variance. The variance is scaled by a². To reverse a coding, X = (Y – b)/a, then apply E(X) = (E(Y) – b)/a and Var(X) = Var(Y)/a². If asked for E(X²), first compute E(Y²) from the coded data, then use the relation E(Y) = aE(X) + b and Var(Y) = a²Var(X) to find E(X) and Var(X), and finally E(X²) = Var(X) + [E(X)]².
纠正： 记住加常数只会平移均值而不影响分散程度，因此方差中的 b 消失。方差缩放因子为 a²。若要反向解码，X = (Y – b)/a，则 E(X) = (E(Y) – b)/a，Var(X) = Var(Y)/a²。如果要求 E(X²)，先由编码数据算出 E(Y²)，然后利用 E(Y) = aE(X) + b 和 Var(Y) = a²Var(X) 求出 E(X) 和 Var(X)，最后 E(X²) = Var(X) + [E(X)]²。
7. Binomial Distribution Conditions | 二项分布的条件
Students often apply the binomial model when the trials are not truly independent or when the probability of success changes. A typical scenario is selecting items without replacement from a small population, where the probability changes after each draw. Another misconception is using the binomial distribution for a variable that counts non-independent occurrences, such as the number of rainy days in a row where weather patterns are dependent.
学生常在试验并非真正独立或成功概率变化的情况下套用二项模型。典型场景是从小总体中不放回抽样，每次抽取后概率都会改变。另一个误区是对于统计非独立发生次数的变量使用二项分布，比如统计连续下雨天数，而天气模式是相互依赖的。
Correction: The binomial model requires a fixed number of trials n, each trial independent, only two outcomes, and a constant probability of success p. When sampling without replacement, use the binomial distribution only if the population is large enough relative to the sample (rule of thumb: population at least 10 times the sample size). Otherwise, the hypergeometric model is appropriate, but WJEC questions will guide you. In exam, look for statements like ‘with replacement’ or ‘assume independence’ before applying a binomial.
纠正： 二项模型要求有固定的试验次数 n、每次试验独立、只有两个结果且成功概率 p 恒定。不放回抽样时，只有当总体相对于样本足够大时才可使用二项分布（经验法则：总体至少是样本量的 10 倍）。否则超几何模型更为适宜，但 WJEC 题目会进行引导。考试中，使用二项分布前要寻找诸如“有放回”或“假设独立性”等表述。
8. Normal Distribution: Continuity Correction | 正态分布：连续性校正
When approximating a binomial distribution with a normal distribution, forgetting to apply the continuity correction is a major error. Students directly standardise the binomial value without adjusting for the discrete to continuous transition, leading to inaccurate probabilities. Another common slip is applying the correction in the wrong direction, e.g. using X ≤ k corrected as Z ≤ (k – 0.5 – μ)/σ instead of adding 0.5.
用正态分布近似二项分布时，忘记进行连续性校正是一个重大错误。学生直接对二项值进行标准化，而不对离散到连续的过渡作调整，导致概率不准确。另一个常见失误是校正方向错误，例如求 X ≤ k 时，错误地将校正写成 Z ≤ (k – 0.5 – μ)/σ 而不是加 0.5。
Correction: For a binomial X ~ B(n, p) approximated by N(np, np(1-p)), use these corrections:
- P(X = k) → P(k – 0.5 < Y < k + 0.5)
- P(X ≤ k) → P(Y < k + 0.5)
- P(X ≥ k) → P(Y > k – 0.5)
Then standardise. Always draw a number line to check the direction. Note that WJEC expects the normal approximation with continuity correction when np and n(1-p) are both greater than 5.
纠正： 对于二项分布 X ~ B(n, p)，用 N(np, np(1-p)) 近似，使用以下校正：
- P(X = k) → P(k – 0.5 < Y < k + 0.5)
- P(X ≤ k) → P(Y < k + 0.5)
- P(X ≥ k) → P(Y > k – 0.5)
然后标准化。始终画个数轴来检查方向。注意 WJEC 要求当 np 和 n(1-p) 都大于 5 时，使用带连续性校正的正态近似。
9. Histograms and Frequency Density | 直方图与频率密度
In a histogram with unequal class widths, the height of each bar must represent frequency density, not raw frequency. A persistent mistake is plotting frequency on the vertical axis for varying widths, which distorts the data representation. Students also struggle to calculate frequency from a histogram: they multiply the bar width by the frequency density but sometimes forget to convert to a consistent unit.
在不等组距的直方图中，每一条的高度必须代表频率密度，而不是原始频数。一个顽固的错误是在宽度不等的组时用纵轴表示频数，这会扭曲数据的呈现。学生还在从直方图计算频数时遇到困难：他们用条宽乘以频率密度，但有时忘记统一单位。
Correction: Frequency density = frequency / class width. When drawing, use frequency density on the y‑axis. To find the frequency of a class, calculate area = class width × frequency density. If the histogram is provided, always measure the width in the units of the x‑axis and multiply by the density (height). It is good practice to annotate the axes clearly as ‘Frequency density’.
纠正： 频率密度 = 频数 ÷ 组距。绘图时，纵轴使用频率密度。要求出一个组的频数，计算面积 = 组距 × 频率密度。如果给出直方图，始终用 x 轴的单位测量宽度，再乘以密度（高度）。最好在轴上清楚标注“频率密度”。
10. Correlation vs Causation | 相关与因果
Many students interpret a strong correlation coefficient as evidence that one variable causes the other. For example, they might conclude that higher ice cream sales cause drowning incidents because both increase in summer. Another error in regression is predicting well outside the range of the given data (extrapolation) and treating the prediction as reliable.
许多学生将强相关系数解释为一个变量导致另一个变量的证据。例如，他们可能得出冰淇淋销量高导致溺水事件增加的结论，因为两者都在夏季上升。回归中的另一个错误是预测远超出给定数据范围（外推），并把该预测视为可靠的。
Correction: Correlation quantifies linear association but does not imply causation. There may be a lurking variable (temperature in the ice cream example) influencing both. When writing about regression, always identify the explanatory (x) and response (y) variables; reversing them will give a different line. Use the regression equation only for interpolation within the observed x‑range. If asked to comment on reliability, note that extrapolation is unreliable.
纠正： 相关系数量化线性关联，但不意味着因果关系。可能存在一个潜伏变量（如上例中的气温）同时影响两者。写回归内容时，始终明确解释变量 (x) 和反应变量 (y)；颠倒它们会得出不同的直线。回归方程只用于观测到的 x 范围内的插值。如果要求评论可靠性，说明外推是不可靠的。
Published by TutorHao | Statistics Revision Series | aleveler.com
更多咨询请联系16621398022（同微信）
屏轩国际教育cambridge primary/secondary checkpoint, cat4, ukiset,ukcat,igcse,alevel,PAT,STEP,MAT, ibdp,ap,ssat,sat,sat2课程辅导，国外大学本科硕士研究生博士课程论文辅导Cancel reply