01 / 21

Confidence Intervals
for a Population Proportion

Learn to construct and interpret a confidence interval for a population proportion (\(p\)), based on the sample proportion \(\hat{p}\).

AP Statistics

What Is a Confidence Interval?

A confidence interval is a range of plausible values for an unknown population parameter. Instead of reporting a single number, we give an interval that we believe contains the true value with a specified level of confidence.

Point Estimate
A single value: our "best guess" from sample data, such as \(\hat{p}\).
Interval Estimate
A range: point estimate \(\pm\) margin of error, capturing uncertainty.

Key Terminology

\(p\)
The true population proportion: the parameter we want to estimate.
\(\hat{p}\)
The sample proportion: our point estimate calculated from data.
\(n\)
The sample size: the number of observations collected.
\(z^*\)
The critical value: depends on the chosen confidence level (e.g., 1.96 for 95%).
Margin of Error
The "buffer" added and subtracted: \(z^*\sqrt{\frac{\hat{p}(1-\hat{p})}{n}}\).

Why Do Confidence Intervals Matter?

A sample proportion \(\hat{p}\) varies from sample to sample. A confidence interval quantifies that variability so decision-makers know how much trust to place in the estimate.

True \(p\)
Sample 1 ✓ Sample 2 ✓ Sample 3 ✕ Sample 4 ✓

The Core Formula

The confidence interval for a population proportion \(p\) is:

\(\hat{p}\)
Center of the interval: your sample proportion.
\(z^*\)
Critical value from the standard normal distribution.
SE
Standard error: \(\sqrt{\frac{\hat{p}(1-\hat{p})}{n}}\).

Critical Values: \(z^*\)

0 -1.96 +1.96 95% 2.5% 2.5%
Confidence Level \(z^*\) Tail Area
90%1.6455% each
95%1.9602.5% each
99%2.5760.5% each

Conditions You Must Check

Before constructing the interval, verify three conditions. If any fails, the interval may not be valid.

1
Random: The data must come from a random sample or randomized experiment.
2
10% Condition (Independence): The sample size \(n\) must be less than 10% of the population: \(n \leq 0.10N\).
3
Large Counts (Normality): Both \(n\hat{p} \geq 10\) and \(n(1-\hat{p}) \geq 10\) must hold.

CI vs. Hypothesis Test

Both are tools for inference about \(p\), but they answer different questions.

Feature Confidence Interval Hypothesis Test
Goal Estimate a range of plausible values for \(p\) Test a specific claim about \(p\)
Uses \(\hat{p}\) in the standard error \(p_0\) in the standard error
Output An interval: \((\text{lower}, \text{upper})\) A p-value and decision
Question "What values of \(p\) are reasonable?" "Is \(p = p_0\) plausible?"

How to Construct the Interval

1
State the parameter: Identify \(p\) in context (e.g., "the proportion of all voters who support the policy").
2
Check conditions: Verify Random, 10%, and Large Counts with calculations shown.
3
Calculate: Compute \(\hat{p}\), find \(z^*\), then evaluate \(\hat{p} \pm z^* \sqrt{\frac{\hat{p}(1-\hat{p})}{n}}\).
4
Interpret in context: "We are [C]% confident that the true proportion of [context] is between [lower] and [upper]."

Example 1: Basic Calculation

A random sample of 200 students finds that 120 prefer online learning. Construct a 95% confidence interval for the true proportion.

1
Sample proportion: \(\hat{p} = \frac{120}{200} = 0.60\)
2
Standard error: \(\text{SE} = \sqrt{\frac{0.60 \times 0.40}{200}} = \sqrt{0.0012} \approx 0.0346\)
3
Margin of error: \(\text{ME} = 1.96 \times 0.0346 \approx 0.0679\)
4
Interval: \(0.60 \pm 0.068 = (0.532,\; 0.668)\)

Visualizing Example 1

0.50 0.55 0.60 0.65 0.70 0.532
\(\hat{p} = 0.60\)
0.668 ME = 0.068 ME = 0.068

Example 2: Checking Conditions

A campaign randomly surveys 500 likely voters in a city of 80,000 and finds 215 support a ballot measure. Construct a 99% CI for the population proportion in favor.

1
Random? Yes: random survey stated. ✓
2
10% Condition? \(500 \leq 0.10 \times 80{,}000 = 8{,}000\). ✓
3
Large Counts? \(n\hat{p} = 215 \geq 10\) and \(n(1-\hat{p}) = 285 \geq 10\). ✓

Example 3: Choosing Sample Size

A researcher wants a 95% CI with a margin of error no more than 0.03. No prior estimate of \(p\) is available. What is the minimum sample size needed?

1
Set up the ME formula: \(z^* \sqrt{\frac{\hat{p}(1-\hat{p})}{n}} \leq 0.03\)
2
Use conservative estimate: \(\hat{p} = 0.5\) (maximizes the product \(\hat{p}(1-\hat{p})\)).
3
Solve for n: Rearrange the ME formula to \(n \geq \hat{p}(1-\hat{p})\left(\frac{z^*}{\text{ME}}\right)^2\). Substitute values: \(n \geq 0.5(0.5)\left(\frac{1.96}{0.03}\right)^2 \approx 1067.1\)
4
Round up: \(n = 1068\) (always round up for sample size).

What Affects the Width?

n = 100 n = 400 n = 1600
\(p = 0.50\)
width ≈ 0.196 width ≈ 0.098 width ≈ 0.049 Quadrupling n halves the width
Increase \(n\)
The interval gets narrower (more precise). Width shrinks by \(\frac{1}{\sqrt{n}}\).
Increase C%
The interval gets wider (more confident). Higher \(z^*\) means larger ME.

Interpreting the Interval Correctly

The interpretation of a confidence interval is one of the most commonly tested concepts on the AP exam. The wording must be precise.

✓ Correct
"We are 95% confident that the true proportion of [context] is between [lower] and [upper]."
✕ Incorrect
"There is a 95% probability that the true proportion is between [lower] and [upper]."

Common Mistakes

✕ Probability Language
Saying "95% probability that \(p\) is in the interval" confuses confidence with probability. Use "confident".
✕ Forgetting Conditions
Skipping the Random, 10%, or Large Counts check loses points on the AP exam and may invalidate the interval.
✕ Using \(p_0\) in SE
For a CI, the standard error uses \(\hat{p}\), not a hypothesized \(p_0\). That formula belongs to hypothesis testing.

Question 1 of 5

In a random sample of 400 adults, 260 said they exercise regularly. What is the sample proportion \(\hat{p}\)?

Question 2 of 5

For a 95% confidence interval, the critical value \(z^*\) is:

Question 3 of 5

Which of the following is a correct interpretation of a 95% confidence interval of \((0.42, 0.58)\)?

Question 4 of 5

A sample of 50 finds \(\hat{p} = 0.12\). Which condition for a confidence interval is not met?

Question 5 of 5

If you want to decrease the width of a confidence interval without changing the confidence level, you should: