Quantiles and Solving for k in IB Mathematics | IB数学：分位数与k值求解方法

📚 Quantiles and Solving for k in IB Mathematics | IB数学：分位数与k值求解方法

Quantiles are fundamental tools in statistics, allowing us to describe the position of a value within a dataset or probability distribution. In IB Mathematics, students frequently need to find a particular value \(k\) such that a given probability condition is satisfied. This article explains the concept of quantiles, explores quartiles and percentiles, and provides a systematic method for solving k-value problems across common distributions.

分位数是统计学中的基础工具，用于描述一个数值在数据集或概率分布中的位置。在IB数学中，学生经常需要找到满足给定概率条件的特定值 \(k\)。本文旨在解释分位数的概念，探讨四分位数和百分位数，并给出在常见分布中求解 k 值的系统方法。

1. What Is a Quantile? | 什么是分位数？

A quantile is a value below which a certain proportion of the data or probability mass falls. For a continuous random variable X, the p-quantile Q(p) is defined by the equation P(X ≤ Q(p)) = p, where 0 < p < 1. In other words, the p-quantile cuts off the lowest p × 100% of the distribution.

分位数是小于或等于某个数值的数据或概率所占比例的对应值。对于连续随机变量 X，p 分位数 Q(p) 由方程 P(X ≤ Q(p)) = p 定义，其中 0 < p < 1。换言之，p 分位数截取了分布最低的 p × 100% 部分。

Common quantiles include the median (p = 0.5), quartiles (p = 0.25, 0.5, 0.75), and percentiles (p = 0.01 through 0.99). In IB exam questions, you may be asked to find a specific quantile or to reverse the process: given a probability, solve for the unknown value k.

常见的分位数包括中位数（p = 0.5）、四分位数（p = 0.25, 0.5, 0.75）和百分位数（p = 0.01 到 0.99）。在IB考试中，题目可能要求你求某个分位数，也可能要求反向操作：给定概率，求解未知量 k。

2. Quartiles and Box-and-Whisker Plots | 四分位数与箱线图

Quartiles split an ordered dataset into four equal parts. The first quartile Q₁ is the 25th percentile, the second quartile Q₂ is the median (50th percentile), and the third quartile Q₃ is the 75th percentile. The interquartile range (IQR) is calculated as Q₃ − Q₁ and measures the spread of the middle 50% of the data.

四分位数将有序数据集分成四个相等部分。第一四分位数 Q₁ 是第25百分位数，第二四分位数 Q₂ 是中位数（第50百分位数），第三四分位数 Q₃ 是第75百分位数。四分位距（IQR）为 Q₃ − Q₁，用于衡量中间50%数据的离散程度。

Measure	Definition	Meaning
Q₁	First quartile	25% of data lies below Q₁
Q₂	Median	50% of data lies below Q₂
Q₃	Third quartile	75% of data lies below Q₃

In a box-and-whisker plot, the box extends from Q₁ to Q₃ with a line at the median. Whiskers extend to the minimum and maximum values within 1.5 × IQR of the quartiles; points beyond this range are considered potential outliers.

在箱线图中，箱子从 Q₁ 延伸到 Q₃，中间线表示中位数。须线延伸到四分位数两侧 1.5 × IQR 范围内的最小值和最大值；超过该范围的点被视为潜在异常值。

3. Percentiles | 百分位数

A percentile is a quantile expressed as a percentage. The kth percentile is the value below which k% of the data lie. For discrete ordered data, there are several conventions for locating percentiles. One standard method uses the position index \(i = \frac{p}{100}(n+1)\), where n is the number of data points and p is the percentile. If i is not an integer, linear interpolation is used.

百分位数是百分数形式表示的分数值。第 k 百分位数是指小于该值的数据占 k%。对于离散的有序数据，有多种定位百分位数的方法。一种标准方法使用位置指标 \(i = \frac{p}{100}(n+1)\)，其中 n 是数据个数，p 是百分位数。如果 i 不是整数，则使用线性插值。

In IB Mathematics, you are generally expected to know how to calculate the median, quartiles, and possibly the 10th and 90th percentiles. Always check whether your calculator or formula sheet uses the \((n+1)\) or \(n\) convention, as results can differ slightly.

在IB数学中，通常要求掌握中位数、四分位数以及可能的第10和第90百分位数的计算。始终检查你的计算器或公式表使用 \((n+1)\) 还是 \(n\) 约定，因为结果可能略有不同。

4. Standard Normal Distribution and z-Scores | 标准正态分布与 z 值

For a normal random variable \(X \sim N(\mu, \sigma^2)\), the standardised value is \(z = \frac{x – \mu}{\sigma}\). The standard normal distribution \(Z \sim N(0, 1)\) has mean 0 and standard deviation 1. The z-score tells us how many standard deviations a value lies from the mean.

对于正态随机变量 \(X \sim N(\mu, \sigma^2)\)，标准化值为 \(z = \frac{x – \mu}{\sigma}\)。标准正态分布 \(Z \sim N(0, 1)\) 的均值为0，标准差为1。z 值表示一个数值距均值多少个标准差。

Most IB problems involving quantiles require converting an unknown k into a z-score. The relationship can be written backwards as \(k = \mu + z_p \cdot \sigma\), where \(z_p\) is the z-score corresponding to the desired percentile p.

大多数涉及分位数的IB题目需要将未知 k 转换为 z 值。这种关系可以反向写成 \(k = \mu + z_p \cdot \sigma\)，其中 \(z_p\) 是所需百分位数 p 对应的 z 值。

5. Inverse Normal Calculation | 反标准正态计算

The inverse normal function, often called invNorm on a calculator, finds the z-score corresponding to a given cumulative probability. For example, to find the value z such that P(Z < z) = 0.95, the calculator returns z ≈ 1.6449. This is the critical value for a one-tailed 5% significance level.

反标准正态函数（计算器上通常称为 invNorm）用于找到与给定累积概率对应的 z 值。例如，要求满足 P(Z < z) = 0.95 的 z，计算器会返回 z ≈ 1.6449。这就是单尾5%显著性水平下的临界值。

When using a normal distribution table, you look up the probability inside the table and read the z-score from the margins. Ensure that the table gives P(Z < z), not P(0 < Z < z), because the two conventions give different values.

使用正态分布表时，在表格内部查找概率，并从边缘读取 z 值。确保表格给出的是 P(Z < z) 而不是 P(0 < Z < z)，因为两种约定给出的数值不同。

Percentile	p	z_p
90th	0.90	1.2816
95th	0.95	1.6449
97.5th	0.975	1.9600
99th	0.99	2.3263

6. General Steps for Solving k | 求解 k 值的一般步骤

When you are asked to find a k-value in a probability context, follow these steps:

当题目要求你在概率情境中求解 k 值时，请按以下步骤操作：

Step 1: Identify the distribution and its parameters.

第1步：识别分布类型及其参数。
Step 2: Write the probability statement, e.g. P(X < k) = 0.8 or P(X > k) = 0.05.

第2步：写出概率语句，例如 P(X < k) = 0.8 或 P(X > k) = 0.05。
Step 3: Adjust the inequality if necessary so that it is in the form P(X ≤ k) = p.

第3步：如有必要，将不等式调整为 P(X ≤ k) = p 的形式。
Step 4: Standardise using \(z = \frac{k – \mu}{\sigma}\) for normal distributions.

第4步：对正态分布使用 \(z = \frac{k – \mu}{\sigma}\) 进行标准化。
Step 5: Use the inverse normal function or statistical tables to find the required z-score.

第5步：使用反标准正态函数或统计表找到所需的 z 值。
Step 6: Solve the resulting equation algebraically for k.

第6步：用代数方法解方程得到 k。

7. Solving for k in a Normal Distribution | 正态分布中的 k 值求解

Let us illustrate the process with an example. Suppose X is normally distributed with mean 100 and variance 225, so X ~ N(100, 15²). We want to find k such that P(X < k) = 0.80.

我们用一个例子来说明这个过程。假设 X 服从均值为100、方差为225的正态分布，即 X ~ N(100, 15²)。我们想要找到满足 P(X < k) = 0.80 的 k。

First, standardise: \(P\left(Z < \frac{k - 100}{15}\right) = 0.80\). From the inverse normal table, the z-score for the 80th percentile is z ≈ 0.8416. Therefore:

首先标准化：\(P\left(Z < \frac{k - 100}{15}\right) = 0.80\)。由反标准正态表，第80百分位数对应的 z 值约为 0.8416。因此：

\(k = 100 + 0.8416 \times 15 = 112.624\)

Thus k ≈ 112.6. This means 80% of values in this distribution are less than 112.6.

因此 k ≈ 112.6。这意味着该分布中80%的值小于112.6。

If the original condition is P(X > k) = 0.05, then P(X ≤ k) = 0.95. Using z₀.₉₅ = 1.6449, we get \(k = 100 + 1.6449 \times 15 = 124.67\). Always remember to convert a right-tail probability into a left-tail probability first.

如果原始条件是 P(X > k) = 0.05，则 P(X ≤ k) = 0.95。使用 z₀.₉₅ = 1.6449，得到 \(k = 100 + 1.6449 \times 15 = 124.67\)。始终记住先要将右尾概率转换为左尾概率。

8. k-Values in Chi-Squared and t-Distributions | 卡方分布与 t 分布中的 k 值

For t-distributions, the notation \(t_{\alpha, \nu}\) represents the value such that the right-tail probability is \(\alpha\) with \(\nu\) degrees of freedom. In hypothesis testing, you often need to find the critical value k. For example, with \(\nu = 10\) and \(\alpha = 0.05\), the critical t-value is \(t_{0.05,10} \approx 1.812\). This value satisfies \(P(T > k) = 0.05\).

对于 t 分布，记号 \(t_{\alpha, \nu}\) 表示自由度为 \(\nu\) 时右尾概率为 \(\alpha\) 的临界值。在假设检验中，通常需要求临界值 k。例如，当 \(\nu = 10\)、\(\alpha = 0.05\) 时，t 临界值为 \(t_{0.05,10} \approx 1.812\)。该值满足 \(P(T > k) = 0.05\)。

For the chi-squared distribution, the notation \(\chi^2_{\alpha, \nu}\) gives the value such that \(P(\chi^2 > k) = \alpha\). Because the chi-squared distribution is not symmetric, you cannot simply flip the sign. Always draw a diagram to identify which tail is involved.

对于卡方分布，记号 \(\chi^2_{\alpha, \nu}\) 表示满足 \(P(\chi^2 > k) = \alpha\) 的临界值。由于卡方分布不是对称的，不能简单地把符号反过来。务必画图来确定涉及的是哪一侧尾部。

Many IB calculators can compute inverse t and inverse chi-squared values directly. In IB Mathematics Analysis and Approaches, these are most commonly used in confidence intervals and goodness-of-fit tests.

许多IB计算器可以直接计算逆 t 和逆卡方值。在IB数学分析与方法中，这些最常用于置信区间和拟合优度检验。

9. Quantiles in a Binomial Distribution | 二项分布中的分位数

For discrete distributions, quantiles are not always exact. The problem is often posed as: find the smallest integer k such that P(X ≤ k) ≥ 0.95, where X ~ Bin(n, p). This k is called the 95th percentile of the distribution.

对于离散分布，分位数不一定精确。题目通常表述为：求最小的整数 k，使得 P(X ≤ k) ≥ 0.95，其中 X ~ Bin(n, p)。这个 k 称为该分布的第95百分位数。

Example: Let X ~ Bin(20, 0.3). We want the smallest k with P(X ≤ k) ≥ 0.95. Using cumulative binomial tables or a calculator, we find:

例：设 X ~ Bin(20, 0.3)。我们要求满足 P(X ≤ k) ≥ 0.95 的最小整数 k。使用累积二项分布表或计算器，我们得到：

\(P(X \leq 8) \approx 0.8866\), \(P(X \leq 9) \approx 0.9520\)

Therefore k = 9. This demonstrates that for discrete distributions, the inequality ≥ is essential, and you must check both sides of the target probability.

因此 k = 9。这说明对于离散分布，不等式 ≥ 是必须的，并且需要检查目标概率两侧的值。

When using the normal approximation to the binomial, apply a continuity correction: if you want P(X ≤ k), use \(P\left(Z < \frac{k + 0.5 - np}{\sqrt{np(1-p)}}\right)\). In the example above, the approximation gives k = 9 as well.

使用正态近似处理二项分布时，需要应用连续性校正：求 P(X ≤ k) 时，使用 \(P\left(Z < \frac{k + 0.5 - np}{\sqrt{np(1-p)}}\right)\)。在上面的例子中，近似同样给出 k = 9。

10. Common Pitfalls and Tips | 常见陷阱与技巧

Students often lose marks on k-value problems due to small but avoidable errors. One common mistake is confusing the direction of the inequality. If the question gives P(X > k), convert it to P(X ≤ k) = 1 − p before using inverse normal.

学生在 k 值题目中常常因为一些细小但可避免的错误失分。一个常见错误是混淆不等式方向。如果题目给出 P(X > k)，请先转换为 P(X ≤ k) = 1 − p，再使用反标准正态。

Another common pitfall is using the wrong degrees of freedom in t or chi-squared problems. For a t-test on a single sample of size n, use \(\nu = n – 1\). In a chi-squared goodness-of-fit test, degrees of freedom depend on the number of categories and estimated parameters.

另一个常见陷阱是在 t 分布或卡方分布问题中使用错误的自由度。对于单个容量为 n 的样本的 t 检验，使用 \(\nu = n – 1\)。在卡方拟合优度检验中，自由度取决于类别数和估计参数的数量。

With multiple-choice or exam questions, always check whether the requested answer is an integer or a continuous value. For binomial distributions, k must be an integer; for normal distributions, k is typically a real number.

在选择题或考试题中，始终检查所求答案是整数还是连续值。对于二项分布，k 必须是整数；对于正态分布，k 通常是实数。

11. Worked Example: Symmetric Interval | 工作示例：对称区间

Consider a random variable X ~ N(50, 8²). Find k such that P(|X − 50| < k) = 0.90.

设随机变量 X ~ N(50, 8²)。求 k 使得 P(|X − 50| < k) = 0.90。

Since the normal distribution is symmetric about the mean, this condition is equivalent to:

由于正态分布关于均值对称，该条件等价于：

\(P(50 – k < X < 50 + k) = 0.90\)

This implies that the total tail probability outside the interval is 0.10, with 0.05 in each tail. Therefore we need the z-score such that P(Z < z) = 0.95, giving z = 1.6449.

这意味着区间外的总尾部概率为 0.10，每条尾部为 0.05。因此我们需要满足 P(Z < z) = 0.95 的 z 值，即 z = 1.6449。

Because the standard deviation is 8, we have k = 1.6449 × 8 = 13.16. So the interval is approximately (36.84, 63.16).

因为标准差为 8，我们有 k = 1.6449 × 8 = 13.16。所以区间约为 (36.84, 63.16)。

This type of problem appears frequently in IB papers because it combines symmetry, inverse normal, and algebraic manipulation in one question.

这类题型经常出现在IB试卷中，因为它将对称性、反标准正态和代数运算结合在同一道题中。

12. Summary | 总结

Quantiles and k-value calculations are central to IB Mathematics statistics. A quantile is simply a value corresponding to a given cumulative probability. To solve for k, identify the distribution, write the probability equation, standardise if needed, and use the inverse function or table to obtain the required value.

分位数和 k 值计算是IB数学统计部分的核心内容。分位数就是与给定累积概率对应的一个值。求解 k 时，应识别分布、写出概率方程、按需标准化，然后利用反函数或表格获得所需值。

Remember the key differences between continuous and discrete distributions. For continuous distributions such as normal and t, the quantile is exact. For discrete distributions such as binomial, the quantile is the smallest integer that satisfies the inequality. With regular practice and attention to tail direction, you can master k-value questions and avoid common traps.

请记住连续分布与离散分布的关键差异。对于正态和 t 分布等连续分布，分位数是精确的。对于二项分布等离散分布，分位数是满足不等式的最小整数。通过定期练习并注意尾部方向，你一定能掌握 k 值题型并避开常见陷阱。

Published by TutorHao | IB Mathematics Revision Series | aleveler.com

更多咨询请联系16621398022（同微信）

Quantiles and Solving for k in IB Mathematics | IB数学：分位数与k值求解方法