Course contentsShow
← Data Science
Lesson 270 of 2,1458. Confidence IntervalsFree lesson

Degrees of Freedom in t-Intervals

Understanding how sample size affects degrees of freedom and the shape of the t-distribution.

Degrees of Freedom in t-Intervals

What you'll learn: How the size of your sample changes the shape of the t-distribution and affects your confidence intervals.

The Core Idea

When you build a confidence interval using the t-distribution (as we just learned), you need to specify something called degrees of freedom (df). For a single sample mean, the degrees of freedom equals your sample size minus one: df = n - 1.

Think of degrees of freedom as "how much independent information" you have. When you calculate a sample mean from n data points, you "use up" one piece of information to estimate that mean. This leaves you with n - 1 independent pieces of information to work with when estimating variability.

Why This Matters for the t-Distribution

Here's the key insight: smaller samples have fewer degrees of freedom, which makes the t-distribution wider and heavier in the tails. This means:

  • Small sample (e.g., n = 5, df = 4): The t-distribution has thick tails and looks quite different from the normal curve. Your critical values are larger, creating wider confidence intervals — reflecting more uncertainty.

  • Large sample (e.g., n = 100, df = 99): The t-distribution looks almost identical to the standard normal distribution. Critical values are smaller, and intervals are tighter.

As your sample size grows, you gain more degrees of freedom, and the t-distribution gradually approaches the normal distribution. This is why the t-distribution is sometimes called the "small sample" distribution — it accounts for extra uncertainty when you don't have much data.

Key Takeaway: Degrees of freedom (n - 1) control the shape of the t-distribution; smaller samples yield fewer degrees of freedom, wider t-distributions, and wider confidence intervals that properly reflect your uncertainty.