📚 Year 11 WJEC Statistics: Common Misconceptions and Corrective Strategies | Year 11 WJEC 统计：常见误区与纠正方法
Statistics is full of subtle traps, and Year 11 students often lose marks on WJEC exams because of deep-rooted misconceptions. Recognising these common errors and applying the right corrective strategies can turn a grade boundary into a secure pass. This article explores the eight most frequent pitfalls in data handling, charts, probability, and inference—and shows you exactly how to overcome them.
统计学中隐藏着许多微妙的陷阱，11 年级的学生在 WJEC 考试中往往因为根深蒂固的误解而丢分。识别这些常见错误并运用正确的纠正方法，可以把一个卡在分数线上下的成绩稳稳地拉上去。本文探讨数据处理、图表、概率和推断中最常见的八个误区，并告诉你具体如何克服它们。
1. Confusing Mean, Median and Mode | 混淆均值、中位数与众数
A widespread mistake is to assume the mean is always the 'best' average, without considering the shape of the data. The mean is heavily influenced by extreme values. For example, in a small company where four employees earn £22 000 and the CEO earns £250 000, the mean salary is £67 600—hardly typical of the workforce.
一个普遍的错误是假定均值总是“最佳”平均数，而不考虑数据的分布形状。均值受极端值的严重影响。例如，在一家小公司里，四名员工的收入为 22 000 英镑，首席执行官的收入为 250 000 英镑，那么平均工资就是 67 600 英镑——这样根本代表不了普通员工的收入水平。
When a distribution is skewed or contains outliers, the median is the more robust measure of central tendency; it resists the pull of extremes. The mode is particularly useful for categorical data or for identifying the most frequent value in discrete data. In a WJEC exam, always justify your choice of average based on the nature of the data, not habit.
当分布呈偏态或包含异常值时，中位数是更稳健的集中趋势度量；它不会受极端值的牵拉。众数对于分类数据或者在离散数据中找出最常出现的值特别有用。在 WJEC 考试中，一定要根据数据的特点来为你选择的平均数提供理由，而不是凭习惯。
2. Misreading Cumulative Frequency Graphs | 误读累积频率图
Many students point at the highest point of a cumulative frequency curve and label it as the median—confusing it with a frequency polygon. In reality, the cumulative frequency graph never drops; its top simply represents the total frequency. To locate the median, you must find the position of the 50th percentile on the vertical axis, draw a horizontal line to the curve, and then drop vertically to the horizontal axis.
许多学生指着累积频率曲线的最高点，把它标为中位数——这是把它和频数多边形混淆了。事实上，累积频率图永远不会下降；它的最高点仅仅代表总频数。要找到中位数，你必须在纵轴上找到第 50 百分位数的位置，画一条水平线与曲线相交，然后垂直向下到达横轴。
Clarify that a cumulative frequency curve’s steepest section corresponds to the interval with the highest frequency density—often the modal class. For quartiles, use the one-quarter and three-quarter positions of the total frequency. This simple ‘halving and quartering’ method removes the guesswork and ensures accuracy with grouped data.
要澄清的一点是，累积频率曲线最陡的部分对应着频数密度最高的区间——通常是众数所在组。至于四分位数，就使用总频数四分之一和四分之三的位置。这种简单的“对半分和四等分”方法可以消除猜测，确保在处理分组数据时准确无误。
3. Misinterpreting Box Plots | 误解箱线图
A common error is to treat the length of the box as an indicator of sample size. In fact, the box represents the interquartile range (IQR), which captures the middle 50% of the data regardless of how many observations there are. A wide box simply means the central half of the data is very spread out.
一个常见错误是把箱体的长度当作样本容量的指标。实际上，箱体代表四分位数距（IQR），它包含了中间 50% 的数据，与观测值的个数无关。箱体宽仅仅意味着数据的中间一半非常分散。
5 × IQR above the upper quartile. Values beyond this are plotted as outliers. Use the box plot to comment on skewness: if the median is closer to the lower quartile and the upper whisker is long, the distribution is positively skewed.
学生们还会把触须误认为是固定的边界线；要记住，在标准的箱线图中，上触须延伸到上四分位数之上 1.5 × IQR 范围内的最大值。超过这个范围的点会被标示为异常值。利用箱线图来判断偏态：如果中位数更靠近下四分位数，且上触须很长，那么分布就是正偏态的。
4. Confusing Histograms with Bar Charts | 混淆直方图与条形图
Drawing a histogram as if it were a bar chart—with equal-width bars and gaps between them—is a classic mistake when data is continuous. A histogram’s bars must touch, and when class widths are unequal, it is the area of each bar that is proportional to frequency, not the height.
在数据是连续的情况下，把直方图画得像条形图——柱子等宽，中间还有间隙——这是一个经典错误。直方图的柱子必须彼此邻接，而且当组距不相等时，柱子的面积与频数成比例，而不是高度。
he formula: frequency density = frequency ÷ class width. Always label the vertical axis as 'Frequency density' ass widths vary. On a WJEC question, you may be asked to complete or interpret a histogram—check the axes and work out class widths before plotting.
正确的做法是使用公式计算频数密度：频数密度 = 频数 ÷ 组距。当组距不等时，始终要把纵轴标注为“频数密度”。在 WJEC 题目中，你可能会被要求补全或解读直方图——在绘图之前一定要检查坐标轴并算好组距。
5. Correlation Does Not Imply Causation | 相关关系不等同因果关系
Spotting a strong positive correlation on a scatter diagram tempts many students to write “X causes Y”. This leaps from statistical association to causal claim without any experimental evidence. Classic examples—like the positive correlation between ice cream sales and drowning incidents—show how a lurking variable such as hot weather can drive both.
在散点图上看到强正相关，许多学生就会忍不住写出“X 导致 Y”。这是在没有实验证据的情况下，从统计关联跳到了因果论断。经典的例子——比如冰淇淋销量和溺水事件之间的正相关——说明了一个潜在变量（如炎热的天气）可以同时驱动两个现象。
In a WJEC response, describe the relationship safely: “As X increases, Y tends to increase, suggesting a positive correlation. However, correlation does not prove causation; other factors may be involved.” This phrasing demonstrates genuine statistical understanding and avoids a common mark-losing trap.
在 WJEC 的答题中，安全地描述关系：“随着 X 增加，Y 往往也增加，这表明存在正相关。然而，相关性并不证明因果关系；其中可能涉及其他因素。”这样的措辞显示出真正的统计理解，避免了常见的丢分陷阱。
6. The Gambler's Fallacy and Independence | 赌徒谬误与独立性
After witnessing a fair coin land heads five times in a row, many students feel tails is now “due”. This gambler’s fallacy overlooks the independence of each toss: the probability of a head on the next flip remains exactly 0.5, no matter what happened before.
在见证一枚公平的硬币连续五次正面朝上之后，许多学生觉得反面“该出来了”。这种赌徒谬误忽视了每次抛掷的独立性：无论之前发生了什么，下一次抛掷出现正面的概率仍然是 0.5。
Reinforce that independent events have no memory. When using tree diagrams for combined events, multiply probabilities along branches for 'AND' outcomes and make sure branch probabilities from a single node sum to 1. For mutually exclusive events, add probabilities for 'OR' situations. Never mix these rules—mistaking 'AND' for 'OR' is another frequent error in WJEC probability questions.
要强化独立事件没有记忆的认知。在使用树形图分析复合事件时，对于“且”的情况，沿分支相乘概率，并确保从同一节点出发的各分支概率之和为 1。对于互斥事件，“或”的情况则概率相加。千万不要把这些规则搞混——把“且”当成“或”是 WJEC 概率题中另一个常见错误。
7. Unrepresentative Samples and Bias | 不具代表性的样本与偏差
Collecting data from the easiest sources—friends, the front row of the classroom, or a social media poll—produces a convenience or voluntary response sample. Such samples are almost always biased because they over-represent people who are easy to reach or who feel strongly about the topic.
从最容易获取的渠道收集数据——比如朋友、教室前排的同学或社交媒体上的投票——产生的就是便利样本或自愿回应样本。这类样本几乎总是有偏差的，因为它们过度代表了容易接触到的人或者对该话题有强烈感受的人。
To make a sample representative, aim for a simple random sample where every member of the target population has an equal chance of being selected. Stratified sampling goes further by ensuring that subgroups appear in the sample in proportion to their size in the population. In exam critiques, explicitly name the type of bias (e.g., selection bias) and explain why the sample is not representative, then propose a better method.
要使样本具有代表性，应该采用简单随机抽样，让目标总体的每个成员都有相等的被选机会。分层抽样更进一步，它确保子群体在样本中的比例与其在总体中的比例一致。在考试中进行批判分析时，要明确指出偏差的类型（例如选择偏差），解释为什么该样本不具有代表性，然后提出更好的方法。
8. Misunderstanding Standard Deviation and Variance | 标准差与方差的误解
A persistent confusion is mixing up standard deviation and variance. Variance is given in squared units, which makes it hard to interpret directly, whereas standard deviation returns to the original measurement unit. Students also often stumble over whether to divide by n or n−1 when calculating a standard deviation.
一个持久的困惑是混淆标准差和方差。方差是用单位的平方表示的，这使得它很难直接解释，而标准差则回到了原始的测量单位。学生在计算标准差时，往往还在为到底除以 n 还是 n−1 而犯难。
In WJEC GCSE Statistics, you will typically deal with sample data, so you should expect to use the sample standard deviation formula which divides by n−1 to give an unbiased estimate of the population standard deviation. If a dataset is explicitly described as the whole population, then division by n is appropriate. Always check the wording of the question. Moreover, avoid comparing standard deviations when the means are very different; a coefficient of variation might be more suitable, though at this level describing the spread in context often suffices.
在 WJEC 的 GCSE 统计中，你通常处理的是样本数据，所以应该预期使用除以 n−1 的样本标准差公式，以给出总体标准差的无偏估计。如果数据集被明确描述为整个总体，那么除以 n 才是合适的。一定要仔细审题。此外，当均值差异很大时，要避免直接比较标准差；变异系数可能更合适，不过在这个层级上，结合背景描述离散程度通常就足够了。
Published by TutorHao | Statistics Revision Series | aleveler.com
更多咨询请联系16621398022（同微信）
或截图后用「扫一扫 → 相册」选择本图