📚 Year 10 OCR Statistics: In-Depth Analysis of Past Papers | 十年级OCR统计：历年真题深度解析
Past papers are the most powerful tool for mastering OCR GCSE Statistics. By analysing real exam questions, patterns emerge that reveal exactly what examiners expect. This article breaks down key topics from Year 10 Statistics, illustrating each with typical past-paper scenarios and step-by-step approaches to help you secure top marks.
历年真题是掌握 OCR GCSE 统计学的最强工具。分析真实考题可以发现命题规律，让你精准把握得分点。本文拆解十年级统计学的关键主题，搭配历年真题常见题型与分步解题法，助你稳拿高分。
1. Data Collection & Sampling | 数据收集与抽样
Past papers frequently begin by testing your ability to identify the population, sample frame and sampling method. A typical question describes a scenario, for example a headteacher surveying students about lunch choices, then asks you to state the population and explain why a sample is used. Examiners expect precise definitions: the population is the entire group of interest, and a sample is a subset selected to represent it.
真题经常先考查你识别总体、抽样框和抽样方法的能力。常见题型是描述一个情景，比如校长调查学生午餐选择，然后要求你说明总体并解释为什么要用样本。考官期望精确定义：总体是所有感兴趣的对象，样本则是选取用于代表总体的子集。
In one common question, students must carry out a stratified sample. You are given a table of year groups with pupil numbers and asked to select a sample of 50. You need to calculate the sampling fraction (50/total) and multiply by each stratum size, rounding appropriately. Always show your working to gain full method marks.
一道常见题要求学生实施分层抽样。题目会给出各年级学生人数表，要求选取容量为50的样本。需要计算抽样比（50÷总人数），然后乘以每个层的大小，适当四舍五入。务必写出计算过程才能拿满方法分。
Examiner's tip: When describing how to use a random number generator or random number tables to pick a simple random sample, give a step-by-step sequence: number the list, generate numbers, ignore repeats, select corresponding individuals.
考官提示：在描述如何使用随机数生成器或随机数表进行简单随机抽样时，给出分步顺序：给名单编号，生成随机数，忽略重复数字，选出对应成员。
2. Interpreting Charts & Diagrams | 图表与示意图解读
sked to calculate an angle or a frequency given total frequency. The formula angle = (frequency / total) × 360° is essential. Many candidates lose marks by misreading the angle from the diagram, so double-check with a protractor if given.
条形图、饼图、茎叶图和累积频率图的题目几乎每年都考。关于饼图，可能要求根据总频数计算角度或频数。公式 角度 = (频数 ÷ 总数) × 360° 至关重要。不少考生因为读错图上的角度而丢分，若提供量角器务必再三核对。
Stem-and-leaf diagrams are used to find median, quartiles and range. A past-paper favourite is to give a back-to-back stem-and-leaf for two groups and ask you to compare their distributions. You must comment on a measure of centre (e.g. median) and a measure of spread (e.g. interquartile range) in context. Simply saying "Group A has a higher median" only scores one mark if you don't also mention what that means in the context of the problem.
茎叶图用于求中位数、四分位数和极差。真题喜欢给出两组的背靠背茎叶图，要求比较它们的分布。你必须结合背景，对集中趋势度量（如中位数）和离散度量（如四分位距）分别给出说明。简单一句"A组的位数更高"只能拿到一分，除非同时说明它在题目情境下的实际意义。
3. Measures of Central Tendency | 集中趋势的度量
grouped data. For grouped data, the mean is estimated using midpoints. A typical question: "Calculate an estimate of the mean height from the grouped frequency table." w the extra column for midpoint × frequency and sum it before dividing by total frequency.
平均数、中位数和众数是许多真题的核心。你需要能从列表、频数表或分组数据中计算它们。对于分组数据，使用组中值估计平均数。典型题目："根据分组频数表估计平均身高。"必须额外列出"组中值×频数"的列，求和后再除以总频数。
Past papers also ask you to choose the most appropriate average. For example, if the data contains an outlier, the median is more representative. Be ready to justify your choice: "The mean is pulled up by a very high value, so the median gives a better typical value." Using statistical vocabulary such as "robust to outliers" can impress examiners.
真题还会让你选择最合适的平均数。例如，如果数据含有异常值，中位数更具代表性。要准备好理由："均值被一个极高值拉高，因此中位数能更好地反映典型值。" 使用"对异常值稳健"这样的统计术语能给考官留下好印象。
Comparing two data sets often requires you to quote both a mean and a standard deviation (or range). When the means are similar, the one with a smaller spread is more consistent. Use comparative language: "The mean of A is greater than that of B, so on average..."
比较两组数据通常需要同时引用平均数和标准差（或极差）。当平均数相当时，离散程度较小的那组更稳定。使用比较性语言："A的均值大于B，因此平均而言……"
4. Measures of Spread: Range and Quartiles | 离散程度的度量：极差与四分位数
Calculating the range, interquartile range (IQR) and percentiles from raw data or stem-and-leaf diagrams is a staple skill. The range is simply maximum – minimum, but it is heavily affected by outliers. The IQR (Q₃ – Q₁) is preferred for skewed data. In an ungrouped list with n values, the position of the median is (n+1)/2. For quartiles, some exam boards accept slightly different methods, but OCR expects you to use the median of the lower half for Q₁ and upper half for Q₃ after finding the median.
从原始数据或茎叶图计算极差、四分位距（IQR）和百分位数是一项基本技能。极差只是最大值减最小值，但极容易受异常值影响。IQR (Q₃ – Q₁) 在偏态数据中更受欢迎。对于含有 n 个值的未分组列表，中位数的位置是 (n+1)/2。四分位数方面，不同考试局方法略有不同，但 OCR 期望你在找到中位数后，用下半部分的中位数作为 Q₁，上半部分的中位数作为 Q₃。
t for the IQR within a larger problem, for instance to identify outliers using the rule Q₁ – el any outliers clearly on a box plot and explain why they are classified as such. Show the calculation in full.
考官常把求 IQR 嵌入到更大的题目中，例如用规则 Q₁ – 1.5×IQR 和 Q₃ + 1.5×IQR 识别异常值。必须在箱线图上清楚标出异常值并解释判定理由。要完整展示计算过程。
5. Box Plots and Cumulative Frequency | 箱线图与累积频率
A standard past-paper task provides a cumulative frequency table or graph and asks you to find the median and quartiles. Reading from the graph: the median corresponds to the 50th percentile, Q₁ at 25% and Q₃ at 75% of the total frequency. Always draw vertical lines down to the axis for each reading and annotate the graph – these marks are explicitly awarded.
历年的标准题型是给出累积频率表或图，要求你找出中位数和四分位数。从图上读值：中位数对应总频数的第50百分位，Q₁ 对应 25%，Q₃ 对应 75%。每次读数都要画垂直线到坐标轴并标注在图上——这些步骤有明确分值。
You may then need to draw a box plot to represent the five-number summary: minimum, Q₁, median, Q₃, maximum. Use a scale and ensure the whiskers are drawn correctly. When comparing two box plots, comment on central tendency (median) and spread (IQR and overall range). The best answers also mention skewness: if the median is nearer to Q₁, the data is positively skewed, etc.
接下来可能需要绘制箱线图来表示五数概括：最小值、Q₁、中位数、Q₃、最大值。使用比例尺并确保触须绘制正确。当比较两个箱线图时，要说明集中趋势（中位数）和离散程度（IQR 和总范围）。优秀答案还会提到偏度：如果中位数更靠近 Q₁，数据为正偏，等等。
6. Probability Basics | 概率基础
A), P(not A), P(A or B) for mutually exclusive events. Remember that for mutually exclusive events, P(A or B) = P(A) + P(B). If the events are independent, P(A and B) = P(A) × P(B). Check carefully whether the question uses replacement or not.
OCR 统计真题的概率题常以双向表或结果列表开篇。你需要找出形如 P(A)，P(非 A)，互斥事件的 P(A 或 B) 的概率。记住：对于互斥事件，P(A 或 B) = P(A) + P(B)。如果事件独立，则 P(A 与 B) = P(A) × P(B)。仔细检查题目是否涉及放回。
An examiner favourite is the "expected frequency" ays in a 60-day period is 0.3 × 60 = 18. Always state that this is an average, not a certainty.
考官偏爱"期望频数"题：用概率乘以试验次数。例如，如果某天下雨的概率是 0.3，那么 60 天中预计下雨天数为 0.3 × 60 = 18。务必指出这只是一种平均，并非必然发生。
7. Probability Tree Diagrams | 概率树图
Tree diagrams are tested heavily. A typical question gives a bag with coloured counters and asks for the probability of drawing two of the same colour, both with and without replacement. You must draw a fully labelled tree, write probabilities on branches, then multiply along branches and add the probabilities of the relevant outcomes. Errors often creep in when probabilities are not written as fractions in their simplest form or when denominators are mixed up.
树图是高频考点。典型题目给出装有彩色计数器的袋子，分别在有放回和无放回的设定下求取到两个同色的概率。你必须画出完整标注的树图，在树枝上注明概率，然后沿枝相乘并将相关结果的概率相加。常见错误是概率没有化为最简分数，或者分母混淆。
appear in higher-tier past papers. You may see a two-event scenario where the second event's probability changes given the first. The notation P(B|A) is used, and OCR often expects you to use the formula P(A and B) = P(A) × P(B|). A clear tree diagram with first and second stage probabilities makes these much easier.
条件概率题出现在高阶真题中。你可能会看到两事件情景，第二个事件的概率会因第一个事件的发生而改变。会用到记号 P(B|A)，OCR 通常期望使用公式 P(A 与 B) = P(A) × P(B|A)。有清晰的带有第一阶段和第二阶段概率的树图后，题目会容易很多。
8. Bivariate Data and Scatter Graphs | 双变量数据与散点图
Scatter graph questions account for a significant portion of past-paper marks. You are given a table of paired values (e.g. hours of revision and test score) and must plot points, describe the correlation, draw a line of best fit, and use it for estimation.
散点图题目在真题中占分不少。你会得到一个配对数据表（例如复习时长和测试分数），需要描点、描述相关性、画出最佳拟合线并用它进行估计。
When describing correlation, use precise language: "strong positive correlation", "weak negative correlation", or "no correlation". Do not just say "positive correlation" unless it is very clear. Also, identify any outliers and comment on their effect. For example, "The outlier (5, 98) lies well above the general trend and may indicate an exceptionally efficient student or a data recording error."
描述相关性时，要用准确的表述："强正相关"、"弱负相关"或"无相关"。不要只说"正相关"，除非非常明显。还要识别异常值并评论其影响。例如："异常点 (5, 98) 远高于整体趋势，可能表明一个学习效率极高的学生，或是数据记录错误。"
9. Correlation and Regression Lines | 相关性与回归线
The line of best fit must be drawn by eye, balancing points above and below. Past-paper marking schemes reward a line that passes through the mean point (mean of x, mean of y) if it has been calculated. When using the line to estimate one value given another, annotations on the graph are vital: draw vertical and horizontal dashed lines and state whether the estimate is interpolation (within the data range) or extrapolation (outside). Extrapolation is less reliable, and you should mention this.
最佳拟合线需要用目测法绘制，使上下方点数大致平衡。真题评分标准奖励穿过均值点 (x 均值, y 均值) 的直线（如果曾计算过）。当使用该直线估计某已知值对应的结果时，图上标注至关重要：画垂直和水平虚线，并说明估计是属于内插（在数据范围内）还是外推（范围外）。外推可靠性较低，应予以说明。
nt m by picking two points on your drawn line (not from the original data unless told) and find the y-intercept c. Use this equation to make predictions, but again caution against extrapolating too far.
一些高阶题目要求写出最佳拟合线的方程 y = mx + c。从你画的直线上取两点（除非题目说明，否则不能直接取原数据点）计算斜率 m，再求截距 c。用此方程做预测，但同样要警惕过度外推。
10. Time Series and Moving Averages | 时间序列与移动平均
Time series graphs show variation over time, and past papers typically ask you to plot the data, then calculate a moving average to smooth out fluctuations and reveal a trend. A 4-point moving average is common with quarterly data. You must average the first four values, then drop the first and add the next, and so on. Plot these moving averages on the same graph to highlight the underlying trend.
时间序列图展示随时间的变化，真题通常要求你先绘制数据点，再计算移动平均值以消除波动、揭示趋势。对于季度数据，4点移动平均较为常见。你需要先平均前四个值，然后去掉第一个、加入下一个，以此类推。将这些移动平均值画在同一张图上以突出潜在趋势。
You might then be asked to predict a future value using the trend line. Extend the moving average line with a dotted line and read off the forecast. Always mention that seasonal effects may not be reflected in this simple trend projection. For full-marks, calculate seasonal variation by subtracting the trend from actual data and express it as an average adjustment.
随后可能要求你用趋势线预测未来值。将移动平均线用虚线延长并读出预测值。务必指出，简单趋势预测无法反映季节性影响。为得满分，可以计算季节变差（实际值减趋势值），并以平均调整量的形式表达。
11. Index Numbers and Ratios | 指数与比率
ast-paper task gives the price of a basket of goods over several years and asks you to calculate chain base index numbers or weighted index numbers. For weighted indices, you also need to multiply each price relative by its weight and divide by total weight.
指数题目常设定一个基年，要求你比较价格变化。公式为 指数 = (当前值 / 基值) × 100。典型真题给出一篮子商品多年的价格，要求计算链基指数或加权指数。对于加权指数，还需将每种价格比乘以各自的权数，再除以总权数。
Ratios appear in resource allocation and scaling questions. For example, "A school's budget is split between departments in the ratio 5:3:2. If the total is £240 000, how much does the first department receive?" Solve by dividing the total by the sum of ratio parts (5+3+2=10) to find one share (£24 000), then multiply by 5. Show units clearly.
比率出现在资源分配和比例缩放题中。例如："一所学校的预算按 5:3:2 的比例分给不同部门。如果总额为240 000英镑，第一个部门分得多少？" 解法是将总预算除以比率份数之和 (5+3+2=10) 得到一份金额 (24 000英镑)，再乘以5。务必清晰标明单位。
12. Common Pitfalls and Exam Tips | 常见错误与应考技巧
One of the biggest exam mistakes is misreading graph scales. Candidates frequently misinterpret a cumulative frequency graph axis as starting from the first class boundary rather than zero, or forget that the plotted point is at the upper class boundary. Always annotate the graph with readings and double-check the scale before answering.
最大失分点是误读图表坐标轴。考生常误以为累积频率图的横轴从第一个组界开始而非零，或忘记描点应在组的上限处。务必在图上标注读数，并在作答前再次确认刻度。
Another frequent error is using the wrong formula for population or sample standard deviation. OCR past papers often provide both options, and you must distinguish between them. For a sample, use the divisor (n-1); for a population, use n. Likewise, when calculating variance, remember to square the deviations from the mean.
另一个常见错误是混淆总体标准差与样本标准差公式。OCR 真题常会同时提供两个选项，你必须加以区分。样本标准差用除数 (n-1)，总体标准差用 n。同样，计算方差时，别忘了先对离均差平方。
Timing is critical. Prioritise questions you find easiest, but always show working – even a correctly substituted formula can earn method marks if the final answer is wrong. Use the mark allocations as a guide to depth: a 1-mark question requires a quick fact or single calculation, whereas a 4-mark comparison question needs detailed, contextualised statements.
时间管理至关重要。优先作答最擅长的题目，但始终展示过程——即使最终答案错误，正确的公式代入也能拿到方法分。以分值为难度指引：1分题只需一个快速事实或单一计算，而4分比较题则需要详细、结合背景的陈述。
Finally, round answers to a sensible degree of accuracy. Unless specified, 1 decimal place more than the original data is a safe rule. If a question states "give your answer to 3 significant figures", adhere strictly – marks are lost for incorrect rounding.
最后，答案要舍入到合理的精确度。除非另有规定，比原始数据多一位小数是安全法则。若题目要求"给出3个有效数字的答案"，必须严格遵守——舍入错误会扣分。
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