📚 Common Mistakes in Year 11 WJEC Statistics and How to Fix Them | Year 11 WJEC 统计常见误区与纠正方法
Even the most confident statistics students can fall into the same traps year after year. For WJEC Year 11 candidates, knowing the content is only half the battle – spotting and sidestepping common errors is what turns a good grade into a great one. This article walks you through the most frequent mistakes, explains why they happen, and shows you exactly how to correct them, with clear examples aligned to the specification.
即使是自信满满的统计学考生，也会年复一年地落入同样的陷阱。对 WJEC Year 11 考生而言，掌握知识只是成功的一半——发现并避开常见错误才能让成绩从良好迈向优秀。本文将带你梳理最常见的误区，解释它们产生的原因，并清晰展示纠正方法，所有例子均紧扣考纲。
1. Misreading Averages: Mean vs Median vs Mode | 误用平均数：均值、中位数与众数
Many students automatically reach for the mean when asked to find an ‘average’, even when the data contains extreme outliers. For example, the salaries in a small company where one person earns £500,000 and everyone else earns under £30,000 will give a mean that does not represent a typical salary. In such cases, the median is far more meaningful because it is not pulled by the extreme value.
许多学生一看到“平均数”就自动计算均值，即使数据中包含极端异常值。例如，一家小公司中一人年薪 500,000 英镑，其余员工都低于 30,000 英镑，此时均值无法代表典型工资水平。在这种情况下，中位数更有意义，因为它不会被极端值拉偏。
The mode, on the other hand, is the only average suitable for qualitative data. A common error is trying to calculate a mean ‘favourite colour’. You cannot add and divide categories, so only the mode makes sense here. Always ask yourself: is my data numerical and symmetric? Use the mean. Is it skewed? Use the median. Is it categorical? Use the mode.
另一方面，众数是唯一适用于定性数据的平均数。常见错误是尝试计算“最喜欢的颜色”的均值。类别无法相加和除法，因此只有众数才有意义。务必自问：我的数据是数值且对称的吗？用均值。数据偏斜吗？用中位数。是分类数据吗？用众数。
2. Correlation Does Not Imply Causation | 相关性不等于因果关系
A classic mistake is to see a strong correlation coefficient, say r = 0.9, between ice cream sales and drowning incidents and then claim that ice cream causes drowning. In reality, a lurking variable – warm weather – drives both. WJEC exam questions often include such scenarios to test your understanding that correlation only measures association, not cause.
一个经典错误是看到冰淇淋销量与溺水事件之间存在强相关系数（如 r = 0.9），便声称冰淇淋导致溺水。实际上，一个隐藏变量——炎热的天气——同时推动了两者。WJEC 考题通常会包含此类情境，以测试你是否明白相关性只度量关联，而非因果关系。
To correct this, always ask: could there be a third factor? When writing exam answers, explicitly state that a high correlation does not prove causation and suggest possible confounding variables. This shows the examiner you are thinking critically, not just calculating.
纠正方法：始终追问，是否存在第三个因素？在答题时，要明确表示强相关不能证明因果关系，并给出可能的混杂变量。这能让考官看到你在进行批判性思考，而非仅仅计算。
3. Misleading Charts and Axes | 图表的误导性坐标轴
Charts that do not start the vertical axis at zero can exaggerate small differences. Students often misinterpret these as showing dramatic changes when, in fact, the variation is tiny. In WJEC papers, you may be asked to explain why a bar chart or line graph is misleading. The usual culprit is an axis that has been truncated or uses an irregular scale.
纵轴不是从零开始的图表会夸大小幅差异。学生常将这些图表误解为巨大变化，而实际波动极小。在 WJEC 试卷中，你可能会被要求解释为什么某张条形图或折线图具有误导性。常见的罪魁祸首是坐标轴被截断或使用了不规则刻度。
To fix your own work, always label axes clearly and, unless there’s a valid reason, start your frequency or value axis at zero. If you must break the axis, indicate it with a zigzag symbol. When critiquing a given graph, calculate the actual percentage change and compare it to the visual impression it gives.
在自己绘图中纠正：始终清晰标记坐标轴，若非有正当理由，频数或数值轴应从零开始。如果必须截断，用锯齿符号标明。在评论所给图表时，应计算出实际的百分比变化，再与图表给人的视觉印象进行对比。
4. Probability Pitfalls: Gambler’s Fallacy | 概率误区：赌徒谬误
After flipping a fair coin and getting five heads in a row, many students believe the next flip is more likely to be tails. This is the gambler’s fallacy – the false belief that past independent events affect future probabilities. In reality, each flip is independent, and the probability of tails remains 0.5, regardless of the previous sequence.
抛一枚公平硬币并连续得到五次正面后，许多学生相信下一次抛出反面的可能性更大。这就是赌徒谬误——错误地认为过去的独立事件会影响未来的概率。实际上，每次抛掷都是独立的，反面的概率始终是 0.5，与之前的结果序列无关。
To avoid this, remind yourself: if events are independent, their probabilities reset each time. In exam questions involving repeated independent trials, never let the history of outcomes alter your probability for the next single trial. Write down the word ‘independent’ on your paper to reinforce the point.
避免这个错误的方法是提醒自己：如果事件相互独立，每次概率都会重置。在涉及重复独立试验的考题中，不要用历史结果更改下一次单次试验的概率。在纸上写下“独立”一词以强化这一点。
5. Tree Diagram Errors: Conditional vs Unconditional | 树状图错误：条件与非条件概率
A very common mistake is to label all branches on a probability tree with unconditional probabilities, even when the situation is without replacement. For instance, if a bag contains 4 red and 6 blue counters and one is removed, the second-branch probabilities must change to reflect the new totals. Using 4/10 and 6/10 again on the second layer is incorrect.
一个非常常见的错误是，在概率树的所有分支上都标记无条件概率，即使情境是不放回的。例如，袋中有 4 个红色和 6 个蓝色筹码，取走一个后，第二层分支的概率必须改变以反映新的总数。再次使用 4/10 和 6/10 就是错误的。
The fix is simple: always ask ‘Does the first event change the sample space?’ If yes, write conditional probabilities like ‘3/9’ and ‘6/9’ on the second branches. And when multiplying along a path, check that you are multiplying the correct conditional probabilities, not mixing them with the first-stage ones inappropriately.
纠正方法很简单：永远要问“第一个事件是否改变了样本空间？”如果改变了，应在第二层分支上写出 3/9、6/9 这样的条件概率。在沿路径相乘时，也要检查是否将正确的条件概率相乘，而没有与第一阶段概率混淆。
6. Sampling Bias: Size Isn’t Everything | 抽样偏差：样本大小并非一切
Students often assume a larger sample is always better, overlooking the critical importance of randomness. A huge but biased sample – for example, surveying only people in a shopping centre at 11 am – will produce worthless results, while a smaller, truly random sample can be reliable. Size does not cure bias.
学生往往认为样本越大越好，却忽略了随机性这一关键因素。一个巨大但有偏差的样本——例如，仅在上午 11 点的购物中心进行调查——会产生毫无价值的结果，而一个较小但真正随机的样本却可能是可靠的。样本大小并不能消除偏差。
To tackle sampling questions, identify the target population and consider whether every member had an equal chance of being selected. If not, point out the sampling method flaw and suggest a simple random or stratified sampling technique. WJEC mark schemes reward specific suggestions over vague ones.
处理抽样问题时，要明确目标总体，并思考是否每个成员都有均等被选中的机会。如果没有，指出抽样方法的缺陷，并建议采用简单随机抽样或分层抽样技术。WJEC 评分标准更青睐具体的建议，而非笼统的说法。
7. Interpreting Standard Deviation | 标准差的误解
A zero standard deviation does not mean the data is useless; it means every value is identical. Some students panic and think they have made an error. Conversely, a very large standard deviation is sometimes dismissed as an anomaly when it actually reveals important variability – for instance, in quality control, high variability might signal a faulty process.
标准差为零并不意味着数据没有用；它意味着每一个数值都相同。有些学生会因此恐慌，以为自己算错了。相反，有时过大的标准差会被当作异常忽略，但它实际上揭示了重要的变异性——例如，在质量控制中，高变异性可能预示着流程存在缺陷。
Remember, standard deviation measures spread around the mean. If two data sets have the same mean but different standard deviations, the one with the larger sd is more spread out. In comparative questions, always pair a comment about the mean with a comment about the standard deviation to give a full picture of both centre and spread.
请记住，标准差衡量的是围绕均值的离散程度。如果两个数据集的均值相同但标准差不同，标准差较大的数据集更分散。在比较性题目中，务必结合均值和标准差进行评论，从而全面呈现数据的中心和离散情况。
8. Box Plot Confusion: IQR and Outliers | 箱线图的困扰：四分位距与异常值
A classic error is calculating the interquartile range (IQR) as Q₃ – Q₁ but then using it incorrectly to find outliers. WJEC uses the rule that an outlier is any value less than Q₁ – 1.5 × IQR or greater than Q₃ + 1.5 × IQR. Many students forget the 1.5 multiplier or apply it only to the maximum and minimum values already in the data, not to the boundary.
一个经典错误是计算出四分位距 IQR = Q₃ – Q₁，但随后却错误地用它来寻找异常值。WJEC 采用的规则是：异常值为任何小于 Q₁ – 1.5 × IQR 或大于 Q₃ + 1.5 × IQR 的数值。许多学生会忘记 1.5 这个乘数，或者只将其应用于数据中已有的最大值和最小值，而非边界。
Another pitfall is interpreting the median line inside the box as the mean. The box plot shows the five-number summary: minimum, Q₁, median, Q₃, maximum. It does not display the mean unless a separate marker is added. Label your diagrams clearly and double-check what each line represents before you describe it.
另一个陷阱是把箱内的中位数线误解为均值。箱线图展示的是五数概括：最小值、Q₁、中位数、Q₃、最大值，除非额外添加标记，它并不显示均值。绘图时要清晰标注，描述之前要再次确认每条线所代表的含义。
9. Histograms: Frequency Density Mistakes | 直方图：频数密度错误
When class widths are unequal, plotting frequency on the vertical axis is a guaranteed way to misrepresent the data. You must use frequency density = frequency ÷ class width. Many students forget this and simply transfer the raw frequencies, making wider intervals visually dominant even when they contain relatively few data points.
当组距不等时，在纵轴上绘制频数必定会歪曲数据。你必须使用频数密度 = 频数 ÷ 组距。许多学生忘记这一点，直接使用原始频数，使得较宽的区间即使在数据点相对较少时也在视觉上占据主导。
In your exam, before drawing anything, check the class widths. If they differ, calculate the frequency densities and label the vertical axis as ‘Frequency density’ not ‘Frequency’. This small habit will prevent a large loss of marks in data presentation questions.
考试中，在绘制前要先检查组距。如果不同，就计算频数密度，并将纵轴标记为“频数密度”而非“频数”。这个小习惯能让你在数据呈现题中避免大量失分。
10. Hypothesis Testing: The p-value Trap | 假设检验：p 值的陷阱
A high p-value (e.g., p = 0.42) does not prove that the null hypothesis is true; it merely indicates that the observed data are consistent with the null hypothesis. Many students incorrectly state ‘the null hypothesis is proved’ or ‘the alternative is false’. In hypothesis testing, we never prove the null – we only fail to reject it.
高 p 值（例如 p = 0.42）并不能证明原假设成立；它仅仅表明观测到的数据与原假设一致。许多学生会错误地声称“原假设得证”或“备择假设为假”。在假设检验中，我们从不证明原假设——只是没有拒绝它。
Furthermore, the significance level (e.g., 5%) is not the probability that the null hypothesis is true. It is the probability of rejecting a true null. To avoid the trap, each time you write a conclusion, use the stock phrase: ‘There is insufficient evidence to reject the null hypothesis’ rather than ‘The null hypothesis is correct’.
此外，显著性水平（如 5%）并不是原假设为真的概率，而是拒绝了正确原假设的概率。为避免陷阱，每次写结论时，都应使用标准表述：“没有足够证据拒绝原假设”，而不是“原假设是正确的”。
11. Time Series: Extrapolation Risks | 时间序列：外推风险
A line of best fit drawn from a time series plot can be extended to predict future values, but students often treat these predictions as certainties. Extrapolation beyond the given data range is risky because trends may change. In WJEC questions, you must qualify any forecast with a statement about the assumption that past trends continue.
由时间序列图绘制的最佳拟合线可以延伸以预测未来数值，但学生常将这些预测视为确定无疑。超过给定数据范围的外推是有风险的，因为趋势可能改变。在 WJEC 题目中，你必须对任何预测进行限定，说明它建立在以往趋势持续这一假设之上。
Also, do not confuse seasonal variation with random fluctuations. If a series shows a peak every fourth quarter, comment on the seasonal pattern, but don’t claim that the next peak is guaranteed. Use the average seasonal effect to adjust your trend forecast, and always mention that real events can override the pattern.
同样，不要混淆季节性变动和随机波动。如果序列每隔四个季度出现一个峰值，可以评论其季节性规律，但不要声称下一个峰值必然会出现。用平均季节效应来调整趋势预测，并始终说明真实事件可能会打破规律。
12. Confusing Percentage Change with Percentage Point Change | 混淆百分比变化与百分点变化
When a pass rate rises from 60% to 75%, the increase is 15 percentage points, but the percentage change is (15 ÷ 60) × 100% = 25%. Many students mix these up, especially when interpreting survey results or exam statistics. This confusion can lead to overstating or understating the true change.
当通过率从 60% 上升到 75% 时，上升的幅度是 15 个百分点，但百分比变化是 (15 ÷ 60) × 100% = 25%。许多学生会把两者混淆，特别是在解读调查结果或考试统计时。这种混淆可能导致夸大或低估真实变化。
To keep them straight, remember: percentage point change is the simple arithmetic difference between two percentages; percentage change is that difference divided by the original percentage. Always pause before you write and ask: am I stating the difference or the relative growth?
要分清它们，请记住：百分点变化是两个百分比之间的简单算术差；百分比变化是该差值除以原始百分比。动笔前先停顿一下，自问：我是在表述差值，还是相对增长？
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