High-Frequency Topics and Common Mistakes in Year 11 CAIE Statistics | Year 11 CAIE 统计高频考点与易错题分析

📚 High-Frequency Topics and Common Mistakes in Year 11 CAIE Statistics | Year 11 CAIE 统计高频考点与易错题分析

In the Year 11 CAIE Statistics course, certain topics appear year after year in examinations, yet students often lose marks on the same predictable pitfalls. Understanding data types, mastering cumulative frequency graphs, interpreting correlation without implying causation, and correctly applying probability distributions are all essential. This article revisits the most tested areas, highlights typical mistakes, and provides practical advice to boost your exam performance.

在 Year 11 CAIE 统计课程中，有些主题年复一年地出现在考试中，但学生们总是在同样的可预见陷阱上丢分。理解数据类型、掌握累积频率图、解释相关关系时不臆断因果，以及正确应用概率分布都是关键。本文回顾了考查最频繁的知识点，指出典型错误，并提供实用建议以提升你的考试成绩。

1. Types of Data and Sampling Methods | 数据类型与抽样方法

The ability to classify data as discrete, continuous, categorical or ordinal is a recurring low-mark question that sets the foundation for later analysis. A classic mistake is labelling shoe size as continuous because half-sizes exist; in reality, sizes take a fixed set of values and are therefore discrete. Similarly, the year a car was manufactured is discrete, not continuous, even though it deals with time.

将数据分类为离散、连续、分类或有序类型是反复出现的低分值题目，却为后续分析奠定基础。经典错误是认为鞋码是连续的，因为存在半码；实际上码数是一个固定值集合，因此属于离散数据。类似地，汽车制造年份也是离散的，并非只因为涉及时间就是连续型。

When choosing a sampling method, students often justify systematic sampling without mentioning the need for a random start. For instance, selecting every 10th person from a register only avoids bias if the first individual is chosen randomly. Another common error is using a convenience sample while claiming it is representative of a population, which invalidates statistical conclusions.

在选择抽样方法时，学生们常常在为系统抽样辩解时忘记提及需要随机起点。例如，从花名册中每隔10人选取一人，只有第一个人是随机选出的才能避免偏差。另一个常见错误是使用便利样本却宣称它能代表总体，这会使统计结论失效。

2. Frequency Tables and Histograms | 频数表与直方图

density = frequency ÷ class width must be applied to every bar, yet many students leave the vertical axis labelled as 'frequency' or use raw frequencies when plotting, which distorts the representation dramatically.

在构建不等宽直方图时，考生常常错误地计算频数密度。公式 频数密度 = 频数 ÷ 组距 必须应用于每一个条形，然而许多学生在纵轴上仍然标注“频数”，或在绘制时直接使用原始频数，这会严重扭曲数据的图形表达。

A typical exam error is forgetting to adjust the frequency density when class intervals are not equal, resulting in bars that mislead the shape of the distribution. For example, a class of width 10 with frequency 20 and a class of width 20 with frequency 30 will have frequency densities 2 and 1.5 respectively; drawing bars of height 20 and 30 would exaggerate the second class.

一个典型的考试错误是当组距不相等时忘记调整频数密度，导致条形的高度歪曲了分布形状。例如，组距为10、频数为20的组与组距为20、频数为30的组，频数密度分别为2和1.5；若绘制高度为20和30的条形，则会夸大第二个组。

Carefully check the scale: the vertical axis must start at zero, and an uneven jump just above zero can misrepresent differences. A common trick in multiple-choice questions is to present a histogram with a broken scale that conceals the true variation – always inspect axis labelling before interpreting.

仔细检查刻度：纵轴必须从零开始，零线以上不规则的跳变会扭曲差异。选择题中常见的陷阱是呈现一个刻度断裂的直方图来隐藏真实变异——在解读前务必检查坐标轴标注。

3. Cumulative Frequency Curves and Percentiles | 累积频率曲线与百分位数

Drawing a cumulative frequency curve seems straightforward, but students often lose marks by plotting points at the upper class boundary, then forgetting to join them with a smooth curve. A common error is using straight line segments, which then gives inaccurate median and quartile readings from the graph.

绘制累积频率曲线看似简单，但学生常因在上界点描点后忘记用平滑曲线连接而失分。常见错误是使用直线段连接，这会导致从图中读取的中位数和四分位数不准确。

orizontal lines from the cumulative frequency axis at the appropriate positions, then dropping vertical lines to the data axis. A careless misreading of the vertical scale, such as taking 0.5 × total frequency instead of 0.5 × (total frequency + 1) for the median position, is a subtle mistake that can change the answer significantly in small datasets.

从曲线上确定中位数（第50百分位数）、下四分位数（第25百分位数）和上四分位数（第75百分位数），需要从累积频率轴的适当位置画水平线，然后向下对数据轴画垂直线。粗心错误如中位数位置的累积频率取 0.5 × 总频数而不是 0.5 × (总频数 + 1)，在小数据集中这种细微错误会显著改变答案。

Candidates sometimes confuse cumulative frequency with cumulative relative frequency. If the question provides relative cumulative frequencies, the median corresponds to 0.5 on the vertical axis, not to half the total frequency. Forgetting to convert back to actual counts can lead to a completely wrong interpretation.

考生有时混淆累积频率与累积相对频率。如果题目给出相对累积频率，那么中位数对应纵轴上的0.5，而不是总频数的一半。忘记转换回实际频数可能导致完全错误的解读。

4. Measures of Central Tendency | 集中趋势的度量

the class width as the multiplier. Always check: midpoint = (lower bound + upper bound) ÷ 2. For open-ended classes, the exam will give an assumed representative value; using your own guess without justification will be penalised.

对于分组数据，估算平均数据需要使用组中点。学生常将上界误当作中点，或使用组距作为乘数。务必核对：中点 = (下界 + 上界) ÷ 2。对于开口组，考试会给出假定的代表值；未经说明而使用自己的猜测会被扣分。

of midpoint × frequency, but then forget to divide by n. For weighted means, it is crucial to multiply each value by its weight before summing, and then divide by the sum of weights.

处理频数表时，高风险错误是忘记最后除以总频数。许多学习者正确地求出中点×频数的乘积之和，却忘记除以 n。对于加权平均数，关键是在求和前将每个值乘以其权重，然后除以权重之和。

The modal class in grouped data is the class with the highest frequency density if class widths are unequal – not simply the one with the highest frequency. This subtle point appears in extended questions and trips up candidates who overlook the distinction between modal class and mode for discrete raw data.

分组数据中的众数组是频数密度最高的组（若组宽不相等）——而不仅仅是频数最高的组。这个细微点出现在拓展题中，会让忽略分组的众数组与离散原始数据众数区别的考生栽跟头。

5. Measures of Dispersion | 离散程度的度量

easure spread, but they serve different purposes. A common mistake is quoting the range as 'UQ – LQ', confusing it with IQR. The IQR = upper quartile – lower quartile, whereas range = maximum – minimum. Always state clearly which measure you are computing.

极差、四分位距（IQR）和标准差各自度量离散程度，但用途不同。一个常见错误是把极差说成“上四分位数减下四分位数”，与 IQR 混淆。IQR = 上四分位数 – 下四分位数，而极差 = 最大值 – 最小值。务必清楚说明你在计算哪种度量。

When calculating standard deviation, check whether the data is a population or a sample. In CAIE Statistics, the population formula uses division by n, while the sample standard deviation uses n – 1. Using the wrong divisor is a marking trap; if the question says ‘for these eight measurements’, treat it as a population unless instructed otherwise. Many exam answers are lost because candidates blindly use the sample formula.

计算标准差时，要核对数据是总体还是样本。在 CAIE 统计中，总体公式除以 n，而样本标准差除以 n – 1。使用错误的除数是评分的陷阱；如果题目说“对于这八次测量”，除非另有说明，应将其视为总体。许多考生因盲目使用样本公式而失分。

For grouped data, the variance formula also uses midpoints, and the same divisor rule applies. Additionally, use exact values from the calculator and round only at the final step; rounding intermediate steps can create an answer that falls outside the accepted tolerance.

对于分组数据，方差公式同样使用中点，且除法规则相同。此外，使用计算器中的精确值，仅在最后一步四舍五入；中间步骤的四舍五入可能使答案超出可接受的容差范围。

6. Box-and-Whisker Plots | 箱线图

nd to the smallest and largest data points that are not outliers. An outlier is typically defined as any value that lies more than 1.5 × IQR below the lower quartile or above the upper quartile. A common misstep is to draw the whisker to the outlier itself, instead of stopping at the adjacent non-outlier value.

根据五数概要做箱线图是最受欢迎的主题之一，但如果须线绘制不正确仍然会丢分。须线延伸到不是异常值的最小和最大数据点。异常值通常定义为低于下四分位数或高于上四分位数超过 1.5 × IQR 的数值。常见失误是将须线画到异常值本身，而不是停在相邻的非异常值上。

When comparing two box plots, avoid vague language like 'box A is higher'. Instead, compare medians explicitly and discuss the IQR and overall range. A typical exam comment requires you to note, for example, that the median of group A is greater than that of group B, and the spread of group B is wider. Missing comparative words means lost communication marks.

比较两个箱线图时，要避免使用“箱 A 更高”这样的模糊语言。应该明确比较中位数，并讨论 IQR 和总极差。典型的考试评述要求你指出，例如，A 组的中位数大于 B 组，且 B 组的分散程度更宽。缺少比较性词语会导致沟通分丢失。

7. Scatter Diagrams and Correlation | 散点图与相关关系

Plotting points accurately is essential, but the most common error on scatter diagrams is describing correlation as 'positive' when points show no clear trend, or confusing correlation with causation. Stating that 'increased ice cream sales cause higher crime rates' because the scatter plot shows an upward trend is a classic fallacy; both may be linked to a third factor such as temperature.

广告 / Sponsored

准确描点至关重要，但散点图最常见的错误是当点无明显趋势时将相关关系描述为“正相关”，或者混淆相关与因果。因散点图呈现上升趋势就声称“冰淇淋销量增加导致犯罪率上升”是经典谬误；二者可能都与第三个因素（如气温）有关。

When asked to describe the correlation, you must mention both strength and direction, e.g., 'strong positive correlation' or 'weak negative correlation'. Simply saying 'there is a correlation' is insufficient. Candidates also forget to identify outliers that lie far from the general pattern; an outlier can strongly influence the correlation coefficient and should be commented upon.

当要求描述相关关系时，必须提及强度和方向，如“强正相关”或“弱负相关”。只说“存在相关”是不够的。考生也常忘记指出远离总体趋势的异常点；异常值会极大地影响相关系数，应当对此加以评述。

Correlation coefficient r always lies between –1 and 1. A value of 0.8 indicates strong positive correlation, while –0.9 indicates strong negative correlation. A common slip is to say r = – n than r = 0.5; the sign indicates direction, so the strength is judged by absolute value. |–0.6| > 0.5, so the negative correlation is actually stronger.

相关系数 r 始终介于 –1 和 1 之间。0.8 表示强正相关，而 –0.9 表示强负相关。常见口误是说 r = –0.6 比 r = 0.5 的相关性更弱；符号表示方向，因此强度由绝对值判断。|–0.6| > 0.5，所以负相关实际上更强。

8. Regression Lines and Predictions | 回归线与预测

ad the intercept a from the output. Be careful: in CAIE Statistics, the line of best fit can be obtained by eye or by calculation, and the formula b = Σ(x – x̄)(y – ȳ) / Σ(x – x̄)² is used for the least-squares regression line.

回归线方程通常以 y = a + bx 给出。学生在计算斜率 b 时经常交换 x 和 y 变量，或从输出结果中误读截距 a。注意：在 CAIE 统计中，最优拟合线可通过目测或计算得到，而最小二乘回归线使用公式 b = Σ(x – x̄)(y – ȳ) / Σ(x – x̄)²。

A very costly error is using the regression line to predict far outside the range of the original data (extrapolation). Predictions for x-values far beyond the given data set are unreliable, and exam questions often specifically test your awareness by asking ‘Explain why this prediction might be inaccurate’. The correct answer must mention extrapolation and that the linear relationship may not hold beyond the observed range.

一个代价巨大的错误是使用回归线去预测远超出原始数据范围的 x 值（外推）。对远离给定数据集的 x 值进行预测是不可靠的，考试题常特意考察你的意识，会问“解释为什么这个预测可能不准确”。正确答案必须提到外推，并说明线性关系可能无法在观测范围之外成立。

When drawing the line of best fit by eye, ensure it passes through the mean point (x̄, ȳ) and balances points on both sides. A common mistake is forcing the line through the origin without justification, which produces a biased model.

当通过目测作最优拟合线时，务必使其穿过均值点 (x̄, ȳ)，并在两侧平衡点数分布。常见错误是毫无依据地强制直线通过原点，这会产生有偏模型。

9. Probability Rules and Tree Diagrams | 概率法则与树形图

ents, P(A and B) = P(A) × P(B). A notorious error is adding probabilities when events can be independent but you need 'or' Maths is 0.6 and Physics is 0.4, and 0.3 study both, then P(Maths or Physics) = 0.6 + 0.4 – 0.3 = 0.7.

概率题考查互斥事件的加法法则 P(A 或 B) = P(A) + P(B) 以及独立事件的乘法法则 P(A 且 B) = P(A) × P(B)。一个常见的易错点是，当事件可能相互独立且需要“或”时简单相加，却不意识到如果事件非互斥则必须减去 P(A 且 B)。例如，一个学生学数学的概率为 0.6，物理为 0.4，两者都学为 0.3，则 P(数学或物理) = 0.6 + 0.4 – 0.3 = 0.7。

Tree diagrams are excellent for conditional probability, but many candidates forget that the probabilities on the second set of branches are conditional on the outcome of the first event. They often reuse the original unconditional probabilites. For example, when drawing without replacement, the probability for the second pick depends on the first outcome; failing to update this causes entirely wrong branch probabilities.

树形图非常适用于条件概率，但许多考生忘记第二组分支上的概率是以第一事件的结果为条件的。他们经常仍用原始的无条件概率。比如，在不放回抽样时，第二次抽取的概率依赖于第一次的结果；未能更新此概率会导致整个分支概率完全错误。

A common pitfall with Venn diagrams is shading the wrong region for intersection or union. When constructing a Venn diagram from a word problem, always fill in the intersection first, then other parts. Reading the question verbatim: 'at least one' means the union, 'both' means intersection, 'neither' is outside both circles.

维恩图中一个常见陷阱是，为交集或并集标错了区域。从文字题构建维恩图时，总是先填交集，再填其它部分。逐字读题：“至少一个”指并集，“两者都”指交集，“两者都不”指在两个圆外。

10. Binomial Distribution | 二项分布

The binomial distribution X ~ B(n, p) applies when there are a fixed number of independent trials, each with only two outcomes (success/failure) and a constant probability of success p. Candidates often misuse the binomial when trials are not independent, such as drawing balls without replacement from a small bag. This requires the hypergeometric situation, but at IGCSE level, if the population is large, binomial approximation may still be used with caution.

二项分布 X ~ B(n, p) 适用于固定次数的独立试验，每次试验只有两种结果（成功/失败），且成功概率 p 恒定。考生经常在试验不独立时误用二项分布，例如从小袋中不放回地抽球。这需要超几何分布，但在 IGCSE 级别，如果总体很大，仍可谨慎使用二项近似。

(X = k). The formula requires the binomial coefficient ⁿCₖ. Many candidates just compute p^k(1 – p)^n–k bability of exactly two heads in three tosses is 3 × (0.5)²(0.5)¹, not simply (0.5)²(0.5).

一个高频错误是在计算 P(X = k) 时忘记考虑组合数。公式需要二项式系数 ⁿCₖ。许多考生只计算 p^k(1 – p)^n–k 而没有乘以方式数。例如，三次抛硬币中恰好两次正面的概率是 3 × (0.5)²(0.5)¹，而不是单纯的 (0.5)²(0.5)。

When finding probabilities such as P(X > 2), use the complement rule where possible. 1 – y: P(X ≥ 3) is not the same as P(X > 3). Check whether the boundary is included or not.

求 P(X > 2) 这样的概率时，尽可能使用补集规则。1 – P(X ≤ 2) 可减少计算量。常见的疏忽是误读不等号：P(X ≥ 3) 与 P(X > 3) 不相同。检查边界是否被包括。

11. Normal Distribution | 正态分布

Standardising a normal variable X to Z is a core skill. The formula Z = (X – ndardise at all and attempt to read probabilities directly from the original distribution.

将正态变量 X 标准化为 Z 是一项核心技能。必须正确应用公式 Z = (X – μ) / σ。许多学生减去均值后除以方差而不是标准差，或者压根忘记标准化而试图直接从原分布查表读取概率。

table requires careful attention to whether the table gives cumulative probabilities from – z). A typical exam error is looking up the wrong tail: for P(Z > a), you must compute 1 – Φ(a). Also, remember that for negative z-values, symmetry gives Φ(–z) = 1 – Φ(z).

使用标准正态表时需要注意表格给出的是从 –∞ 到 z 的累积概率，还是从 0 到 z 的。CAIE 的表格通常代表 Φ(z) = P(Z < z)。典型的考试错误是查错了尾部：对于 P(Z > a)，你必须计算 1 – Φ(a)。此外，要记住对于负 z 值，利用对称性 Φ(–z) = 1 – Φ(z)。

In 'backwards' n looking up the z-table. For instance, if P(Z > k) = 0.2, then P(Z < k) = 0.8, so look up 0.8 inside the table to find k. Students often look up 0.2, obtaining the wrong sign.

在“反向”问题中，给定一个概率需要求对应的 z 值或 X 值。常见错误是查 Z 表时忽视了概率的方向。例如，若 P(Z > k) = 0.2，则 P(Z < k) = 0.8，因此应在表内查找 0.8 以得到 k。学生常常查找 0.2，从而得到错误的符号。

12. Exam Strategy and Common Pitfalls | 考试策略与常见陷阱

Beyond content, time management and reading depth are crucial. In statistics, context is everything: a question about 'mean time taken' expects units in the answer. Leaving off units, or using the wrong degree of rounding, loses accuracy marks. Always check the required precision: 'to the nearest kg', '3 significant figures'.

除内容外，时间管理和读题深度也至关重要。统计中的语境就是一切：关于“所花平均时间”的问题，答案中必须有单位。漏写单位或使用了错误的舍入精度会导致精确度分数丢失。始终检查所需精度：“精确到整数 kg”、“3 个有效数字”。

When answering a 'compare' or 'comment' question, use statistical vocabulary precisely. Phrase like 'the median of A is higher, suggesting...' and support with figures. Aim for two comparison points, one about location (mean/median) and one about spread (IQR/range). Vague answers like 'they are different' do not score.

回答“比较”或“评述”类问题时，要精确使用统计术语。用例如“A 的中位数更高，这表明……”这样的句式，并用数据支撑。力争写出两个比较点：一个关于位置（平均数/中位数），一个关于离散度（IQR/极差）。模糊答案如“它们不一样”是无法得分的。

Finally, practice past paper questions under timed conditions, specifically focusing on topics that involve multi-step calculations, such as grouped standard deviation or cumulative frequency interpretations. Revise your errors methodically – the most common mistakes are rarely about complex theory, but about skipping steps or misreading the demands of the question.

最后，在限时条件下练习历年真题，特别关注涉及多步计算的主题，如分组标准差或累积频率解读。有条理地复习你的错误——最常见的失误很少是因为理论复杂，而是因为跳步或误读题目要求。

Published by TutorHao | Statistics Revision Series | aleveler.com

更多咨询请联系16621398022（同微信）