01 / 21

Justifying Claims:
Confidence Intervals for \(\mu\)

Learn how to use a confidence interval to make justified claims about a population mean and connect interval estimation to hypothesis testing.

AP Statistics

What Is a Confidence Interval?

A confidence interval is a range of plausible values for an unknown population parameter, computed from sample data. For a population mean \(\mu\), the interval takes the form:

where \(\bar{x}\) is the sample mean, \(s\) is the sample standard deviation, \(n\) is the sample size, and \(t^*\) is the critical value from the \(t\)-distribution.

Anatomy of the Interval

\(\bar{x}\)
Lower Bound Upper Bound Margin of Error
\(\bar{x}\): Point Estimate
The center of the interval: your best single guess for \(\mu\).
Margin of Error
\(t^* \cdot \dfrac{s}{\sqrt{n}}\): how far the interval extends in each direction.

What Does 95% Confidence Mean?

A 95% confidence level means: if we repeated the sampling process many times and built a confidence interval each time, about 95% of those intervals would capture the true population mean \(\mu\).

True \(\mu\)
Misses \(\mu\)
Captures \(\mu\)

How to Justify a Claim About \(\mu\)

To use a confidence interval to justify a claim about \(\mu\), apply this simple decision rule:

1
Construct (or be given) a confidence interval for \(\mu\).
2
Check whether the claimed value \(\mu_0\) lies inside or outside the interval.
3
If \(\mu_0\) is inside the interval: the data are consistent with the claim. We cannot reject it.
4
If \(\mu_0\) is outside the interval: we have convincing evidence against the claim at that confidence level.

The CI-Test Duality

A confidence interval and a two-sided hypothesis test are two sides of the same coin. A value \(\mu_0\) falls outside a \(C\%\) confidence interval if and only if a two-sided test of \(H_0\!: \mu = \mu_0\) would reject at significance level \(\alpha = 1 - C/100\).

⇔ 95% CI
\(\mu_0\) outside interval
→ Reject claim Two-sided test
\(\alpha = 0.05\)
\(p < 0.05\) → Reject \(H_0\)
Same decision, different framing

Conditions for a Valid \(t\)-Interval

Before constructing or interpreting a confidence interval for \(\mu\), verify these three conditions:

Random
Data come from a random sample or randomized experiment.
Normal/Large
Population is Normal, or \(n \geq 30\) (CLT), or graph shows no strong skew/outliers.
Independent
Observations are independent. If sampling without replacement: \(n \leq 10\%\) of population.

CI vs. Hypothesis Test

Confidence Interval Hypothesis Test
Purpose Estimate a range for \(\mu\) Test a specific claim about \(\mu\)
Output An interval \((L, U)\) A p-value and decision
Reject when \(\mu_0\) falls outside CI \(p\text{-value} < \alpha\)
Direction Always two-sided Can be one- or two-sided
Extra info Shows all plausible values Only addresses one \(\mu_0\)

Step-by-Step Approach

1
Identify the claim. What value of \(\mu\) is being claimed? Call it \(\mu_0\).
2
Check conditions. Random, Normal/Large, Independent (10% condition).
3
Construct the interval. \(\bar{x} \pm t^* \cdot \frac{s}{\sqrt{n}}\) at the given confidence level.
4
Compare. Is \(\mu_0\) inside or outside the interval?
5
State your conclusion in context. "Because \(\mu_0\) is (not) in the interval, we do (not) have convincing evidence that..."

Example 1: Coffee Shop Wait Times

A coffee shop claims the mean wait time is 5 minutes. A random sample of \(n = 36\) customers gives \(\bar{x} = 5.8\) min and \(s = 2.4\) min. A 95% CI is constructed:

4.988 6.612
\(\bar{x} = 5.8\)
\(\mu_0 = 5\)

Example 2: Battery Life Claim

A manufacturer claims their batteries last 500 hours. A consumer group tests \(n = 50\) batteries and finds \(\bar{x} = 486\) hrs, \(s = 40\) hrs. They build a 95% CI:

474.63 497.37
\(\bar{x} = 486\)
\(\mu_0 = 500\)

Example 3: Changing Confidence Level

Using the same battery data (\(\bar{x} = 486\), \(s = 40\), \(n = 50\)), what happens if we use a 99% confidence interval instead?

470.84 501.16
\(\bar{x} = 486\)
\(\mu_0 = 500\)
99% CI: wider interval, \(\mu_0\) now inside

Writing a Justification

On the AP exam, you must connect the interval to the claim in context. Use this template:

1
State the interval: "The 95% confidence interval for the mean [context] is \((L, U)\)."
2
Locate the claim: "The claimed value of [value] is (inside / outside) this interval."
3
Conclude: "Therefore, we (do / do not) have convincing evidence that the true mean [context] is different from [value]."

Interpreting Confidence Level Correctly

✓ Correct
"We are 95% confident that the interval \((L, U)\) captures the true population mean."
✕ Incorrect
"There is a 95% probability that \(\mu\) is in this interval." (\(\mu\) is fixed: it either is or isn't.)

Common Mistakes to Avoid

✕ Probability Language
Saying "95% chance \(\mu\) is in \((L, U)\)." The parameter is fixed; probability applies to the method.
✕ Ignoring Conditions
Constructing a CI without checking Random, Normal/Large, or Independent conditions.
✕ Interval for Data
Stating that the interval contains 95% of the sample data. The interval is an estimate for the population mean, not individual data points.

Summary Reference

Scenario CI Finding Conclusion
\(\mu_0\) inside CI Plausible value Do not reject the claim
\(\mu_0\) outside CI Not plausible Reject the claim at level \(\alpha\)
Increase confidence level Wider interval Harder to reject (more conservative)
Increase sample size \(n\) Narrower interval More precise estimate of \(\mu\)
\(\mu_0\) near boundary Borderline Collect more data or raise \(n\)

Question 1 of 5

A 95% confidence interval for the mean weight of cereal boxes is (15.6, 16.4) ounces. The manufacturer claims the mean weight is 16 ounces. What is the correct conclusion?

Question 2 of 5

A 90% confidence interval for mean commute time is (22.3, 28.7) minutes. A city official claims the mean commute is 30 minutes. What can we conclude?

Question 3 of 5

Which statement is a correct interpretation of a 95% confidence interval?

Question 4 of 5

A 95% CI for mean test score is (72.1, 79.9). A 99% CI from the same data is (70.4, 81.6). A teacher claims \(\mu = 80\). Which is true?

Question 5 of 5

A researcher constructs a confidence interval for mean blood pressure from a voluntary response sample of 200 patients. The interval is (118, 126). What is the biggest concern?