AQA 数学 a level further maths example responses paper 4 unit fs2 对比 | AQA Maths A-Level Further Maths Example Responses Paper 4 Unit FS2 Comparison

--- title: AQA进阶数学FS2高分答题范例与考点解析 subtitle: AQA Further Maths FS2 Example Responses & Exam Tips author: TutorHao date: 2026-08-08 cover: /mnt/d/网站项目/jd-union/data/covers/cover_02_b.png --- ## 一、FS2考卷结构与分值分布 | FS2 Paper Structure & Mark Distribution AQA A-Level Further Mathematics Paper 4 Unit FS2（Further Statistics 2）是进阶数学中统计方向的核心模块。整张试卷满分50分，考试时间1小时30分钟，通常包含5至6道大题，每道大题下设若干小问。 FS2 paper carries 50 marks with a 1 hour 30 minute time limit. It typically consists of 5 to 6 multi-part questions, each progressing from basic calculation to higher-order interpretation. The assessment objectives (AOs) are split roughly as: AO1 (use and apply standard techniques) 50%, AO2 (reason, interpret and communicate mathematically) 25%, AO3 (solve problems within mathematics and in other contexts) 25%. The key topics tested in FS2 include the Poisson distribution and its approximations, Chi-squared tests (goodness of fit and association), continuous random variables (probability density functions, cumulative distribution functions, expectation and variance), linear combinations of independent random variables, confidence intervals, and hypothesis testing for population parameters. Understanding the command words in questions — "state", "find", "show that", "interpret", "comment on" — is essential for targeting your answer at the right level. FS2的评分标准非常注重答题步骤的完整性。即使最终答案错误，只要方法正确且步骤清晰，仍可获得方法分（M marks）和准确性跟进分（A1 ft marks）。因此，在答题时务必写出完整的计算过程，不要跳步。 ## 二、泊松分布条件检验与概率计算 | Poisson Distribution Conditions & Probability Calculations 泊松分布是FS2的第一个重点。考试中常见的题型是：给出一个真实场景，要求判断是否满足泊松分布条件，然后计算特定事件的概率。 The Poisson distribution models the number of events occurring in a fixed interval of time or space. For a Poisson distribution to be valid, three conditions must hold: 1. **Events occur singly** — two events cannot happen at exactly the same instant. 2. **Events occur at a constant average rate** — the mean number of occurrences per unit interval, λ, is constant. 3. **Events occur independently** — the occurrence of one event does not affect the probability of another. In exam questions, always state these conditions explicitly when asked "State the conditions under which a Poisson distribution would be a suitable model." A common mistake is to write "events are random" — this is too vague and will not earn the mark. Use the precise phrasing above. When calculating probabilities, remember that for X ~ Po(λ): P(X = r) = (e^(-λ) × λ^r) / r! For cumulative probabilities such as P(X ≥ k), convert to 1 - P(X ≤ k-1) and use the statistical tables provided in the exam. Always show this conversion step clearly in your working. **Example response (scoring full marks):** *Question: "A call centre receives calls at an average rate of 4.2 per minute. Find the probability that more than 5 calls are received in a randomly chosen one-minute interval."* *Model answer:* Let X be the number of calls received in one minute. X ~ Po(4.2). P(X > 5) = 1 - P(X ≤ 5) From tables, P(X ≤ 5) = 0.7479 (using λ = 4.2). Therefore, P(X > 5) = 1 - 0.7479 = 0.2521 (4 d.p.) 注意：在从泊松分布表查值时，如果λ不是整数，应使用线性插值。这一技巧经常出现在FS2考试中。例如，当 λ = 4.2 时，可以查找 λ = 4.0 和 λ = 4.5 的值进行插值计算。 ## 三、泊松分布近似二项分布的判定与应用 | Poisson Approximation to Binomial: Conditions & Applications When n is large and p is small, a binomial distribution B(n, p) can be approximated by a Poisson distribution Po(np). The rule of thumb is: n > 50 and np < 5, or simply "n is large, p is small." The exam will often ask you to "suggest a suitable approximating distribution" before calculating probabilities. Your answer must include: 1. The original distribution: X ~ B(n, p) 2. The condition check: n is large (state n explicitly), p is small (state p explicitly) 3. The approximating distribution: X ≈ Po(np) 4. The calculated probability using the Poisson tables **Common pitfall:** Students sometimes approximate without checking conditions. Always verify n and p against the criteria and write the justification in your answer. AQA examiners specifically look for the phrase "since n is large and p is small." **Example response:** *Question: "3% of light bulbs produced by a factory are defective. A random sample of 200 bulbs is selected. Find the probability that at most 4 bulbs are defective."* Let X be the number of defective bulbs. X ~ B(200, 0.03). Since n = 200 is large (> 50) and p = 0.03 is small (< 0.1), we can approximate: X ≈ Po(λ = np = 200 × 0.03 = 6) P(X ≤ 4) = 0.2851 (from Poisson tables with λ = 6) If the question also asks for the exact binomial probability, note that the Poisson approximation gives 0.2851 while the exact binomial value is 0.2856. The difference is negligible, demonstrating the quality of the approximation. ## 四、卡方拟合优度检验完整步骤 | Chi-Squared Goodness of Fit: Complete Step-by-Step The Chi-squared goodness of fit test (χ² GOF test) is a high-mark question in FS2, typically worth 8-12 marks. You must present a full, structured answer. **Standard 7-step response template:** 1. **State the hypotheses** H₀: The data follows the specified distribution / The model is a good fit. H₁: The data does not follow the specified distribution / The model is not a good fit. 2. **Calculate expected frequencies** For each category, Eᵢ = n × P(category). If any Eᵢ < 5, combine adjacent categories. 3. **State the significance level and degrees of freedom** ν = number of categories after combining - number of estimated parameters - 1 4. **Find the critical value from χ² tables** State the critical value clearly with the significance level. 5. **Calculate the test statistic** χ² = Σ (Oᵢ - Eᵢ)² / Eᵢ. Show the full calculation table. 6. **Compare and conclude** If χ²_calc > χ²_crit: reject H₀. Otherwise: do not reject H₀. 7. **Interpret in context** State what your conclusion means in the context of the original problem. **Example response (AQA-style):** *Question: "A die is rolled 120 times with observed frequencies: 1→26, 2→15, 3→22, 4→18, 5→19, 6→20. Test at the 5% level whether the die is fair."* H₀: The die is fair (p₁ = p₂ = ... = p₆ = 1/6) H₁: The die is not fair (at least one pᵢ ≠ 1/6) Expected frequency for each face: Eᵢ = 120 × 1/6 = 20 (all Eᵢ ≥ 5, no combining needed) ν = 6 - 0 - 1 = 5 Critical value: χ²₅(5%) = 11.070 (from tables) χ² = (26-20)²/20 + (15-20)²/20 + (22-20)²/20 + (18-20)²/20 + (19-20)²/20 + (20-20)²/20 = 36/20 + 25/20 + 4/20 + 4/20 + 1/20 + 0/20 = 1.8 + 1.25 + 0.2 + 0.2 + 0.05 + 0 = 3.50 Since 3.50 < 11.070, we do not reject H₀. There is insufficient evidence at the 5% level to suggest the die is unfair. 注意：当期望频数小于5时，必须合并相邻类别，并相应调整自由度。这是一个非常容易丢分的地方，很多学生合并了类别但忘记调整ν。 ## 五、卡方独立性检验与列联表分析 | Chi-Squared Test for Association & Contingency Tables The Chi-squared test for association (also called test for independence) uses a contingency table. The core difference from the GOF test is in the calculation of expected frequencies and degrees of freedom. For an r × c contingency table: - Expected frequenc y: Eᵢⱼ = (row total × column total) / grand total - Degrees of freedom: ν = (r - 1)(c - 1) **Key exam technique:** Always check that all expected frequencies are ≥ 5. If this condition is not met, you must state that "the Chi-squared test may not be valid" and suggest either combining rows/columns or using Fisher's exact test (though the latter is beyond the AQA syllabus, mentioning it shows broader understanding). **Example response scoring full marks:** *Question: "A survey asks 200 students whether they prefer online or in-person learning, grouped by year group (Year 12, Year 13). Test at the 1% significance level whether preference is associated with year group."* Observed frequencies: | | Online | In-person | Total | |-----------|--------|-----------|-------| | Year 12 | 45 | 55 | 100 | | Year 13 | 63 | 37 | 100 | | Total | 108 | 92 | 200 | H₀: Preference and year group are independent. H₁: Preference and year group are associated. Expected frequencies (row total × column total ÷ grand total): E₁₁ = 100 × 108 / 200 = 54 E₁₂ = 100 × 92 / 200 = 46 E₂₁ = 100 × 108 / 200 = 54 E₂₂ = 100 × 92 / 200 = 46 All expected frequencies ≥ 5, so the test is valid. ν = (2 - 1)(2 - 1) = 1 Critical value: χ²₁(1%) = 6.635 χ² = (45-54)²/54 + (55-46)²/46 + (63-54)²/54 + (37-46)²/46 = 81/54 + 81/46 + 81/54 + 81/46 = 1.5 + 1.761 + 1.5 + 1.761 = 6.522 Since 6.522 < 6.635, we do not reject H₀. There is insufficient evidence at the 1% level to suggest an association between preference and year group. Note: Always comment on the validity of the test by checking expected frequencies before proceeding with calculations. This earns a specific mark. ## 六、连续随机变量PDF与CDF的互推计算 | Converting Between PDF and CDF for Continuous Random Variables Continuous random variables are a core FS2 topic where students frequently lose marks due to algebraic errors or missing domain restrictions. For a continuous random variable X with probability density function f(x): - The cumulative distribution function F(x) = P(X ≤ x) = ∫₋∞ˣ f(t) dt - The median m satisfies F(m) = 0.5 - The mode is the value of x that maximises f(x) — find by differentiating and setting f'(x) = 0 **Critical rules for full marks:** 1. Always state the domain of F(x) as a piecewise function: 0 for x below the support, the integral expression for x within the support, and 1 for x above the support. 2. When finding the median, set F(m) = 0.5 and solve. Only accept solutions within the support. 3. The expectation E(X) = ∫ x·f(x) dx over the entire support. E(X²) = ∫ x²·f(x) dx. Then Var(X) = E(X²) - [E(X)]². **Example response:** *Question: "A continuous random variable X has probability density function f(x) = kx(4-x) for 0 ≤ x ≤ 4, and 0 otherwise. (a) Find k. (b) Find the cumulative distribution function F(x). (c) Find the median of X."* (a) Since ∫₀⁴ f(x) dx = 1: ∫₀⁴ kx(4-x) dx = k ∫₀⁴ (4x - x²) dx = k [2x² - x³/3]₀⁴ = k [2(16) - 64/3] = k [32 - 64/3] = k [96/3 - 64/3] = k × 32/3 = 1 Therefore k = 3/32. (b) For 0 ≤ x ≤ 4: F(x) = ∫₀ˣ (3/32)t(4-t) dt = (3/32) ∫₀ˣ (4t - t²) dt = (3/32)[2t² - t³/3]₀ˣ = (3/32)(2x² - x³/3) = (3x²/16) - (x³/32) So F(x) = { 0, x < 0; (3x²/16) - (x³/32), 0 ≤ x ≤ 4; 1, x > 4 } (c) F(m) = 0.5 → (3m²/16) - (m³/32) = 0.5 Multiply by 32: 6m² - m³ = 16 → m³ - 6m² + 16 = 0 By trial: m = 2 gives 8 - 24 + 16 = 0 ✓ So (m - 2)(m² - 4m - 8) = 0 m = 2 or m = 2 ± 2√3 ≈ 2 ± 3.464 Only m = 2 and m = 2 + 2√3 lie in [0, 4]. Check F(2) = 0.5 ✓, F(5.464) is outside [0,4]. Therefore median = 2. ## 七、独立随机变量线性组合的期望与方差 | Linear Combinations of Independent Random Variables This topic tests your understanding of expectation and variance operators. The key formulas are: For independent random variables X and Y, and constants a, b: - E(aX ± bY) = aE(X) ± bE(Y) - Var(aX ± bY) = a²Var(X) + b²Var(Y) — always addition for variance! For a sum of n independent and identically distributed (i.i.d.) random variables: - E(X₁ + X₂ + ... + Xₙ) = nE(X) - Var(X₁ + X₂ + ... + Xₙ) = nVar(X) **The Central Limit Theorem (CLT):** For a large sample size (n ≥ 30), the sample mean X̄ is approximately normally distributed with mean μ and variance σ²/n, regardless of the original distribution. **Example response (FS2 style):** *Question: "The weight of apples from an orchard is normally distributed with mean 150g and standard deviation 20g. A crate contains 36 randomly selected apples. Find the probability that the total weight of the apples exceeds 5.5kg."* Let Wᵢ be the weight of the i-th apple. Wᵢ ~ N(150, 20²). Let T = W₁ + W₂ + ... + W₃₆ be the total weight. E(T) = 36 × 150 = 5400g Var(T) = 36 × 20² = 36 × 400 = 14400 Standard deviation(T) = √14400 = 120g So T ~ N(5400, 120²). P(T > 5500) = P(Z > (5500 - 5400) / 120) = P(Z > 0.8333) = 1 - Φ(0.8333) = 1 - 0.7977 = 0.2023 The probability that the total weight exceeds 5.5kg is approximately 0.202. A common error is dividing the variance instead of multiplying — remember: Var(nX) = n²Var(X) but Var(X₁ + ... + Xₙ) = nVar(X) when the variables are independent. These are numerically different situations and the exam will test your ability to distinguish them. ## 八、FS2高分答题策略与考场时间分配 | FS2 Exam Strategy & Time Management After analyzing multiple AQA examiner reports, here are the most impactful strategies for maximising your FS2 score: **1. "方法分优先"策略 | Method Marks First** Even if you cannot complete a calculation, write down the correct method. For example, if you forget the Poisson formula but remember P(X > 5) = 1 - P(X ≤ 5), write that conversion step — it earns a method mark independent of the final answer. **2. 假设检验的完整结构 | Complete Hypothesis Test Structure** Always write: Hypotheses → Significance level → Distribution under H₀ → Test statistic → Critical value → Comparison → Conclusion in context. Missing any element loses marks. The conclusion must reference both the significance level and the context. **3. 时间分配建议 | Time Allocation** With 50 marks in 90 minutes, aim for roughly 1.8 minutes per mark. For a typical 10-mark Chi-squared question, allocate 18 minutes. This leaves 12 minutes for checking. Never spend more than 5 minutes stuck on one part — move on and return later. **4. 计算器使用技巧 | Calculator Skills** Your calculator can compute Poisson and binomial probabilities directly. Use it to verify table readings, but always show the "1 - P(X ≤ k)" conversion step in your written working — the examiner needs to see your reasoning. **5. 常见的表述扣分点 | Common Wording Penalties** - "Accept H₀" → Wrong! Say "do not reject H₀" (我们无法拒绝原假设) - "The data proves..." → Wrong! Say "there is evidence to suggest..." (数据表明...) - Writing P(X = x) as just "probability" without specifying the distribution → Always write the full notation. 掌握这些答题技巧后，配合充足的真题练习，FS2取得高分并非难事。建议至少完成2018年至今的全部AQA FS2真题，重点分析评分方案中的给分点分布。 ---

📚 需要A-Level进阶数学辅导？

专业导师一对一辅导 · 真题精讲 · 考点突破

📞 16621398022（同微信）

访问 aleveler.com 了解更多课程信息

更多咨询请联系16621398022（同微信）