📚 Year 11 AQA Statistics: High-Frequency Topics and Common Mistake Analysis | Year 11 AQA 统计高频考点与易错题分析
As Year 11 students prepare for the AQA GCSE Statistics exam, identifying the most frequently assessed topics and understanding the common pitfalls can significantly boost confidence and marks. This revision guide highlights key areas such as sampling, histograms, standard deviation, correlation, and probability, while drawing attention to mistakes that examiners see year after year. Read on to strengthen your statistical reasoning and avoid losing easy grades.
随着 Year 11 学生备战 AQA GCSE 统计考试，明确高频考点并理解常见失分点能显著提升信心与分数。这篇复习指南聚焦于抽样、直方图、标准差、相关和概率等关键领域，同时着重分析那些评分者年年都会看到的典型错误。阅读下去，加强你的统计推理能力，避免丢分。
1. Sampling Methods: Avoiding Stratified Sampling Errors | 抽样方法：避免分层抽样错误
take is to calculate the sample size per stratum incorrectly. For example, if a school has 200 teachers, 800 support staff and 1000 students, and a sample of 110 is required, the number of teachers in the sample should be (200/2000) × 110 = 11. Students often forget to multiply by the total sample size or use the wrong denominator.
分层抽样要求将总体分为不同的层，并按比例从每层中选取样本。一个常见错误是错误计算每层的样本量。例如，如果一所学校有200名教师、800名后勤人员和1000名学生，需要抽取110个样本，那么教师样本数应为 (200/2000) × 110 = 11。学生经常忘记乘以总样本量，或者使用了错误的分母。
- Forgetting to multiply the stratum proportion by the overall sample size.
忘记用层比例乘以总样本量。 - Using an incorrect denominator when finding the proportion (e.g. total stratum size instead of total population).
计算比例时用错分母（如用该层总数代替总体总数）。 - Not rounding sensibly – sample sizes for each stratum must be integers and should sum to the required total.
未合理取整 – 每层样本量必须为整数且之和应等于所需总数。
2. Questionnaire Design: Pitfalls in Data Collection | 问卷设计：数据收集的陷阱
A well-designed questionnaire is crucial for collecting reliable data. Common errors include using leading questions (e.g. 'Don't you agree that school meals are healthy?'), overlapping response boxes in tick-box questions, and not providing an exhaustive set of options. If a question asks 'How old are you?' with categories '0-10, 10-20, 20-30', a 10-year-old or 20-year-old would fit into two categories – a classic boundary overlap mistake.
一份设计良好的问卷对于收集可靠数据至关重要。常见错误包括使用引导性问题（例如“你难道不认为学校膳食很健康吗？”）、选择题选项之存在重叠区间，以及未提供穷尽的选项集合。如果一个问题询问“你的年龄是多少？”，类别为“0-10, 10-20, 20-30”，那么10岁或20岁的人会落入两个类别 – 这是一个典型的界限重叠错误。
- Using leading or biased wording that pushes the respondent towards a particular answer.
使用带有引导性或偏向性的措辞，将受访者推向某个特定答案。 - Allowing overlapping numerical categories or non-mutually exclusive options.
允许数值类别重叠，或选项不互斥。 - Omitting an 'Other' or 'Prefer not to say' option where necessary, forcing respondents to pick an inaccurate answer.
在必要时遗漏“其他”或“不愿回答”选项，迫使受访者选择不准确的答案。
3. Histograms and Frequency Density: Getting the Scale Right | 直方图与频率密度：正确掌握尺度
common mistake is plotting raw frequency on the vertical axis, which distorts the data. For example, if the interval 0 ≤ x < 20 has frequency 30 and 20 ≤ x < 30 has frequency 25, the first bar should have height 30/20 = 1.5, while the second bar height is 25/10 = 2.5. Students who simply plot 30 and 25 will give a misleading picture.
在组距不等时，直方图依据的是频率密度而非频率。公式为频率密度 = 频率 ÷ 组距。最常见的错误是在纵轴直接标画频数，这会扭曲数据。例如，若区间 0 ≤ x < 20 的频数为30，而 20 ≤ x < 30 的频数为25，则第一条柱高应为 30/20 = 1.5，第二条柱高为 25/10 = 2.5。直接画30与25的学生会呈现一幅误导性的图表。
- Forgetting that area represents frequency – height alone is meaningless in histograms with varying widths.
忘记面积代表频率 – 在不等宽直方图中，单独的柱高无意义。 - Incorrectly calculating class width – 20 the width is 11 (or 10 depending on context); always check if boundaries are inclusive.
错误计算组距 – 例如区间 10 ≤ x ≤ 20 的宽度可能是11（或根据情境为10）；始终确认边界是否包含。 - Confusing frequency density with frequency per unit, and not labelling the vertical axis clearly.
混淆频率密度与单位频数，并未清楚标记纵轴。
4. Cumulative Frequency and Box Plots: Avoiding Median and Quartile Missteps | 累积频率与箱线图：避免中位数和四分位数失误
nd quartiles. A common error is misreading the graph: the lower quartile corresponds to 25% of the total frequency, the median to 50%, and the upper quartile to 75% le: lower fence = Q1 − 1.5 × IQR, upper fence = Q3 + 1.5 × IQR. Forgetting to multiply by 1.5 or using the range instead of IQR are frequent slip-ups.
累积频率图用于估计中位数和四分位数。一个常见错误是读图不当：下四分位数对应总频率的25%，中位数对应50%，上四分位数对应75%。学生们有时会在数据轴上取一半的点，而不是在频率轴上。在箱线图中，异常值采用1.5 × IQR规则识别：下围栏 = Q1 − 1.5 × IQR，上围栏 = Q3 + 1.5 × IQR。忘记乘以1.5，或使用极差而非IQR，都是经常出现的疏漏。
- Drawing the cumulative frequency curve through points plotted at the upper class boundary instead of using a smooth curve.
累积频率曲线绘制时穿过上组界的点后未连成光滑曲线。 - Reading the median from the cumulative frequency axis incorrectly – it is found at half the total frequency, not half the maximum data value.
错误地从中累积频率轴读取中位数 – 应在总频率一半处查找，而非数据最大值的一半。 - 1.0 × IQR instead of 1.5 × IQR to locate outliers, causing genuine outliers to be missed.
使用1.0 × IQR而非1.5 × IQR来定位异常值，导致真正的异常值被遗漏。
5. Standard Deviation: Population vs Sample (n vs n−1) | 标准差：总体与样本（n与n−1）
tion or just a sample. For a population, σ = √(Σ(x − μ)² / N). For a sample, s = √(Σ(x − x̄ e population formula for sample data or forget to square root the final result altogether.
标准差衡量数据的离散程度。公式取决于你拥有的是整个总体还是仅仅一个样本。对于总体，σ = √(Σ(x − μ)² / N)。对于样本，s = √(Σ(x − x̄)² / (n − 1))。对样本使用n而非n−1会低估真实的离散程度。许多考生错误地将总体公式用于样本数据，或者完全忘记对最终结果开平方。
s = √( Σ(x − x̄)² / (n−1) )
- Confusing σ (population) and s (sample) when the context asks for 'an unbiased estimate of the population standard deviation'.
当题目要求“总体标准差的无偏估计”时混淆了σ（总体）与s（样本）。 - Subtracting the mean incorrectly or squaring negative deviations without parentheses – calculator entry errors are common.
减去均值时出错，或在平方负偏差时未加括号 – 常见的计算器输入错误。 - Failing to take the square root at the end, and presenting variance instead of standard deviation.
最后未能开平方，给出了方差而非标准差。
6. Scatter Graphs and Correlation Coefficients: Interpreting Strength and Type | 散点图与相关系数：解读强度与类型
ming a strong correlation based on a graph with a clear pattern but many outliers, or stating 'causal relationship' when only correlation is evident. When interpreting Pearson' 0.8 indicates strong positive correlation, but r = 0 does not always mean no relationship – the relationship could be non-linear. Students often mislabel correlation as causation.
散点图直观显示两个变量之间的关系。相关性可以是正、负或零。一个常见错误是根据一张虽有清晰形态但包含诸多异常点的图声称存在强相关性，或者在仅见相关时宣称存在“因果关系”。解读皮尔逊积矩相关系数r时，0.8表明强正相关，但r = 0并不总意味着无关系 – 关系可能为非线性的。学生们经常将相关性误标为因果性。
- Drawing a line of best fit that does not pass through the mean point (x̄, ȳ) or ignoring outliers that pull the line.
绘制的最佳拟合线未通过均值点(x̄, ȳ)，或者忽略了会拉动直线的异常值。 - Using a point on a scatter graph for extrapolation without acknowledging the danger of unreliable predictions beyond the data range.
使用散点图上的点进行外推，却未意识到超出数据范围的预测不可靠的危险。 - Confusing the correlation coefficient with the slope of the regression line – a steep slope can still have a weak correlation if points are widely scattered.
混淆相关系数与回归线的斜率 – 如果点很分散，陡峭的斜率仍可能仅有弱相关性。
7. Spearman's Rank: Handling Ties and Ranking Correctly | 斯皮尔曼等级相关：正确处理并列数据
Spearman's rank correlation coefficient rs is used when data is non-parametric or ordinal. The formula rs s share the same value and would occupy ranks 4 and 5, each must be assigned the average rank (4+5)/2 = 4.5. Simply assigning 4 and 5 is wrong and alters d². Also, forgetting to square differences d or misusing the formula with n³−n in the denominator (which is equivalent to n(n²−1)) can cost marks.
斯皮尔曼等级相关系数rs用于非参数或有序数据。公式 rs = 1 − (6Σd²)/(n(n²−1)) 依赖于正确排序的数据。最常见的错误发生在有并列值时。例如，若两个数据点取值相同且应占据第4与第5名，则每个须赋予平均秩次 (4+5)/2 = 4.5。简单地赋予4和5是错误的，会改变d²。此外，忘记将差值平方或分母误用 n³−n（这与 n(n²−1) 等价），都可能丢分。
- Ranking in the wrong order – ensure both sets are ranked consistently (e.g. both from highest to lowest).
排序方向错误 – 确保两组排秩一致（例如都从最高到最低）。 - Using Spearman's formula on untied data without checking for correlation first – Spearman's rs only measures monotonic relationships.
在未先检验相关性的情况下对未捆绑数据使用斯皮尔曼公式 – 斯皮尔曼rs仅度量单调关系。 - Simply subtracting one rank from the other without handling ties, leading to an inflated Σd².
仅将一个秩次减去另一个而未处理并列情况，导致Σd²被高估。
8. Normal Distribution and z-scores: Standardising Correctly | 正态分布与z分数：正确标准化
ted to z-scores: z = (x − μ) / σ. A typical mistake is using the sample standard deviation s instead of the population σ, or swapping x and μ in the numerator. When reading from the standard normal table, students sometimes read the complementary probability incorrectly – y may read Φ(1.5) as the tail. Always sketch the distribution and shade the required area.
正态分布是一条以均值μ和标准差σ定义的钟形对称曲线。求概率时，须将数据化为z分数：z = (x − μ) / σ。一个典型错误是使用样本标准差s代替总体σ，或在分子中交换x与μ。查阅标准正态分布表时，学生们有时会错误地读取互补概率 – 例如P(Z > 1.5)是1 − Φ(1.5)，但他们可能把Φ(1.5)当成尾部面积。始终绘制分布草图并给所需区域涂上阴影。
- Assuming data is normally distributed without checking, or applying z-scores to a clearly skewed dataset.
未经检验就假设数据呈正态分布，或将z分数用于明显偏斜的数据集。 - Not linearising a question – if asked for the interquartile range, standardise Q1 and Q3 separately before finding the difference.
未将问题线性化 – 若要求四分位距，须分别标准化Q1和Q3再求差。 - Using the wrong table or confusing Φ(z) with the probability density f(z).
使用了错误的表，或将Φ(z)与概率密度f(z)混淆。
9. Probability Trees and Conditional Probability: Untangling 'Given That' | 概率树与条件概率：理顺“已知”条件
f branches using unconditional probabilities instead of conditional ones. For example, if a bag contains 3 red and 5 blue, the first draw probabilities are 3/8 and 5/8. After removing one red, the second red probability becomes 2/7, not 3/8. Conditional probability questions often require P(A|B) = P(A∩ ly branch probabilities without dividing by P(B), or confuse which event is the condition.
概率树图有助于可视化多阶段试验。从一个节点伸出的枝条概率之和必须为1。一个常见错误是在编写第二层枝条时使用了非条件概率，而非条件概率。例如，若袋中有3红5蓝，首次抽取概率为3/8和5/8。在移去一个红球后，第二次抽到红的概率变为2/7，而非3/8。条件概率题常需用到P(A|B) = P(A∩B)/P(B)。许多考生错误地将枝条概率相乘而未除以P(B)，或混淆了哪一事件是条件。
- Not reducing fractions or simplifying early, leading to arithmetic errors in the final answer.
未约分或过早简化，导致最终答案出现算术错误。 - Labelling branches with cumulative probabilities instead of conditional probabilities, making the entire tree invalid.
将累加概率而非条件概率标注在枝干上，导致整个树无效。 - Forgetting that probabilities on a tree are multiplied along branches and added across different paths – misapplying the 'or' rule.
忘记树图中沿枝干相乘、跨路径相加 – 误用了“或”规则。
10. Binomial Distribution: Mean and Variance Calculations | 二项分布：均值与方差计算
ame probability of success p, the number of successes X follows a binomial distribution: X ~ is using the mean formula for variance or vice versa. When calculating P(X = k) = C(n,k) pk (1−p)n−k, students often miscalculate the binomial coefficient C(n,k) or forget to include (1−p)n−k. In problems asking for 'at least' or 'at most', summing several probabilities is required – candidates sometimes only calculate the single value P(X = k).
当固定次数的独立试验每次具有相同的成功概率p时，成功次数X服从二项分布：X ~ B(n, p)。期望成功次数为μ = np，方差为σ² = np(1−p)。一个常见错误是将均值公式用于方差，或反过来。计算P(X = k) = C(n,k) pk (1−p)n−k时，同学们经常算错二项式系数C(n,k)，或忘记包含(1−p)n−k。在要求“至少”或“至多”的问题中，需要将几个概率相加 – 考生有时只计算单个值P(X = k)。
- Using p and 1−p the wrong way round – if p is given as 0.2, then 1−p is 0.8, not 0.2.
将p与1−p弄反 – 如果给定p为0.2，那么1−p为0.8，而非0.2。 - Treating trials as independent when they are not (e.g. without replacement), which requires a hypergeometric approach instead.
在试验不独立时（如不放回抽样）仍视为独立，此时应改用超几何分布。 - Stating np as the variance: variance = np(1−p); the standard deviation is √(np(1−p)).
将np说成方差：方差 = np(1−p)；标准差为 √(np(1−p))。
11. Statistical Diagrams: Choosing the Right Graph | 统计图表：选择正确的图示
Selecting the appropriate diagram is a skill tested frequently in the AQA exam. Bar charts are for categorical data, histograms for continuous data with unequal intervals, pie charts for proportions of a whole, and stem-and-leaf diagrams for small sets of quantitative data while preserving original values. A common mistake is using a line graph for discrete unrelated categories, or drawing a bar chart when a histogram with frequency density is needed. Furthermore, pictograms must use symbols of consistent size and scale. Pupils occasionally misinterpret a cumulative frequency curve as representing exact values rather than enabling estimates.
选择合适的图表是AQA考试中经常考察的技能。条形图适用于分类数据，直方图适用于组距不等连续数据，饼图用于显示整体中比例，茎叶图则用于小型量化数据集并能保留原始数值。一个常见错误是对不相关的离散类别使用折线图，或在需要频率密度直方图时绘制了条形图。此外，象形图必须使用尺寸与比例一致的符号。学生偶尔会误将累积频率曲线解读为代表精确数值，而非用于估算。
- Choosing a dual bar chart for comparing two data sets when component parts are needed – a stacked bar chart or back-to-back stem-and-leaf may be better.
当需要显示组成部分时，却选择了双条形图来对比两组数据 – 此时堆叠条形图或背靠背茎叶图可能更好。 - Drawing a line graph for discrete data and connecting unrelated points – use vertical line charts for discrete frequency.
绘画散点连接的折线图用于离散数据，将不相关点连接 – 离散频数应使用垂线图。 - Not including a key or appropriate labels, making the diagram useless despite correct construct.
未包含图例或适当标签，致使图表虽构造正确却无法解释。
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