Confidence Intervals for a Population Mean with Unknown Variance (Data) | 方差未知时总体均值的置信区间（数据）

📚 Confidence Intervals for a Population Mean with Unknown Variance (Data) | 方差未知时总体均值的置信区间（数据）

When the population variance is unknown, constructing a confidence interval for the population mean requires the t-distribution. This article explains the theory, outlines the step-by-step calculation, and works through a data example.

当总体方差未知时，为总体均值构造置信区间需要使用 t 分布。本文将解释相关理论、逐步计算方法，并通过一个数据示例进行演示。

1. Population Mean and Variance: Known vs Unknown | 总体均值与总体方差：已知与未知

Statistical inference often focuses on the population mean μ. If the population variance σ² is known, a z-interval based on the standard normal distribution is appropriate. If σ² is unknown, we must estimate it from the sample using the sample variance s², which introduces extra uncertainty.

统计推断通常关注总体均值 μ。如果总体方差 σ² 已知，可以使用基于标准正态分布的 z 区间。如果 σ² 未知，我们只能利用样本方差 s² 来估计它，这会引入额外的不确定性。

The sample mean x̄ is an unbiased estimator of μ, and the sample standard deviation s is an estimator of σ. Because s varies from sample to sample, the width of the confidence interval must account for this variation. Using the t-distribution achieves this automatically.

样本均值 x̄ 是 μ 的无偏估计量，样本标准差 s 是 σ 的估计量。由于 s 会随样本不同而变化，置信区间的宽度必须考虑这一波动。使用 t 分布可以自动实现这一要求。

2. Why Use the t-Distribution | 为什么使用 t 分布

The t-distribution is used when the population variance is unknown because it accounts for the extra variability introduced by estimating σ with s. As a result, t-intervals are generally wider than z-intervals with the same confidence level, especially for small samples.

当总体方差未知时使用 t 分布，是因为它考虑了用 s 估计 σ 所带来的额外变异。因此，在相同置信水平下，t 区间通常比 z 区间更宽，在小样本时尤为明显。

As the sample size n increases, the t-distribution approaches the standard normal distribution. This means the difference between the t and z critical values becomes negligible for large n, but for small n the correct t critical value is essential for accurate intervals.

随着样本量 n 增大，t 分布趋近于标准正态分布。这意味着当 n 很大时，t 与 z 临界值之间的差异可以忽略，但在小样本时，使用正确的 t 临界值对于构造准确的区间至关重要。

3. The t-Distribution and Degrees of Freedom | t 分布与自由度

The shape of the t-distribution is controlled by the degrees of freedom, abbreviated df. For a single-sample inference about the mean, the degrees of freedom are df = n − 1, where n is the sample size.

t 分布的形状由自由度 df 控制。对于关于均值的单样本推断，自由度 df = n − 1，其中 n 是样本量。

The degrees of freedom represent the number of independent pieces of information available in the sample. The t-distribution has thicker tails when df is small, reflecting more uncertainty; as df increases, it looks increasingly like the normal distribution.

自由度代表样本中可用的独立信息量。当 df 较小时，t 分布尾部更厚，体现更大的不确定性；随着 df 增大，它越来越接近正态分布。

4. The Confidence Interval Formula | 置信区间公式

The general form of a confidence interval for a population mean when σ² is unknown is given by:

当 σ² 未知时，总体均值的置信区间一般形式为：

x̄ ± t* · (s / √n)

In this formula, x̄ is the sample mean, s is the sample standard deviation, n is the sample size, and t* is the critical t-value with n − 1 degrees of freedom that puts an area of α/2 in the upper tail. For a 95% confidence level, α = 0.05, and the area in each tail is 0.025.

在该公式中，x̄ 是样本均值，s 是样本标准差，n 是样本量，t* 是具有 n − 1 个自由度、使得上侧尾部面积为 α/2 的临界 t 值。对于 95% 置信水平，α = 0.05，每侧尾部面积为 0.025。

5. Step-by-Step Procedure | 构建置信区间的步骤

To construct a confidence interval for the population mean with unknown variance, follow these steps.

当方差未知时，为总体均值构造置信区间的步骤如下。

Check the conditions: the sample must be a simple random sample, and either the population distribution is approximately normal or the sample size is large enough for the Central Limit Theorem to apply.

检查条件：样本必须是简单随机样本，且总体分布近似正态，或者样本量足够大从而使中心极限定理成立。
Calculate the sample mean x̄ and the sample standard deviation s from the data.

根据数据计算样本均值 x̄ 和样本标准差 s。
Choose the confidence level, for example 95%. Determine α = 1 − confidence level, and find the critical t* with df = n − 1 and tail area α/2.

选择置信水平，例如 95%。确定 α = 1 − 置信水平，并查找 df = n − 1、尾部面积为 α/2 的临界值 t*。
Compute the margin of error E = t* × (s / √n).

计算误差幅度 E = t* × (s / √n)。
Write the confidence interval as x̄ − E < μ < x̄ + E, or equivalently (x̄ − E, x̄ + E).

将置信区间写为 x̄ − E < μ < x̄ + E，或等价地 (x̄ − E, x̄ + E)。

6. Worked Example: 95% Confidence Interval from Sample Data | 数值示例：由样本数据计算 95% 置信区间

Suppose a teacher records the test scores of 10 randomly selected students from a large class: 68, 72, 75, 78, 80, 82, 83, 85, 88, 91. The population variance is unknown, so we will construct a 95% confidence interval for the mean test score.

假设一位老师记录了一个大班中随机抽取的 10 名学生的测验成绩：68, 72, 75, 78, 80, 82, 83, 85, 88, 91。总体方差未知，因此我们为平均成绩构造 95% 置信区间。

The sample mean is x̄ = 80.2. The sample standard deviation is calculated using the formula s = √[ Σ(xᵢ − x̄)² / (n − 1) ] = √(459.6 / 9) ≈ 7.15.

样本均值为 x̄ = 80.2。样本标准差按公式 s = √[ Σ(xᵢ − x̄)² / (n − 1) ] 计算，得到 s = √(459.6 / 9) ≈ 7.15。

For a 95% confidence level and df = 9, the critical t-value is t* = 2.262. The margin of error is E = 2.262 × (7.15 / √10) ≈ 2.262 × 2.26 ≈ 5.11.

对于 95% 置信水平和 df = 9，临界 t 值为 t* = 2.262。误差幅度为 E = 2.262 × (7.15 / √10) ≈ 2.262 × 2.26 ≈ 5.11。

Therefore, the 95% confidence interval for the population mean is 80.2 ± 5.11, giving (75.09, 85.31).

因此，总体均值的 95% 置信区间为 80.2 ± 5.11，即 (75.09, 85.31)。

7. Using a Calculator or GDC | 使用计算器或 GDC

Many graphing calculators have a built-in t-interval function. For example, on a TI-84 calculator, press STAT, choose TESTS, then select TInterval. You can enter the data list or the summary statistics x̄, s, and n.

许多图形计算器内置了 t 区间功能。例如，在 TI-84 计算器上，按 STAT，选择 TESTS，然后选 TInterval。你可以输入数据列表，或输入汇总统计量 x̄、s 和 n。

After choosing the confidence level, the calculator outputs the confidence interval directly. This is a quick validation tool, but understanding the underlying formula and assumptions is still essential for the IB exam.

选择置信水平后，计算器会直接输出置信区间。这是快捷的验证工具，但理解背后的公式和假设对于 IB 考试仍然至关重要。

8. Interpreting Confidence Intervals and Confidence Level | 解释置信区间与置信水平

A 95% confidence interval does not mean the probability that μ lies inside this particular interval is 95%. Instead, it means that if we repeated the sampling process many times and constructed the interval each time, approximately 95% of those intervals would contain the true population mean.

95% 置信区间并不表示 μ 落在这个特定区间内的概率为 95%。它的含义是：如果多次重复抽样过程并每次都构造区间，那么大约 95% 的区间会包含真实的总体均值。

Thus, we say we are “95% confident” that the true mean lies between the lower and upper bounds of the calculated interval. This interpretation is a common exam question, so it must be expressed carefully.

因此，我们说“有 95% 的置信度”认为真实均值位于所计算区间的上下限之间。这种解释是常见的考点，需要表述准确。

9. Assumptions and Common Mistakes | 假设与常见错误

The t-interval relies on three key assumptions: independent random sampling, an approximately normal population distribution (or a large enough sample), and no extreme outliers. Violating these assumptions can make the interval unreliable.

t 区间依赖三个关键假设：独立随机抽样、总体分布近似正态（或样本量足够大）、以及没有极端异常值。违反这些假设会使区间不可靠。

Common mistakes include using a z-value instead of a t-value, confusing σ with s, using the wrong degrees of freedom, or misinterpreting the confidence level. For example, using z* = 1.96 instead of t* = 2.262 in the example above would produce a narrower and incorrectly precise interval.

常见错误包括：使用 z 值而不是 t 值、混淆 σ 和 s、使用错误的自由度、或错误解释置信水平。例如，在上面的例子中用 z* = 1.96 代替 t* = 2.262 会得到更窄但不正确的区间。

10. Summary | 总结

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

更多咨询请联系16621398022（同微信）

Confidence Intervals for a Population Mean with Unknown Variance (Data) | 方差未知时总体均值的置信区间（数据）