01 / 21

Justifying Claims:
Confidence Intervals

Learn to construct and interpret confidence intervals for a population proportion and use them to justify statistical claims.

AP Statistics

What Is a Population Proportion?

A population proportion \( p \) is the fraction of individuals in an entire population that have a particular characteristic. Since we rarely survey everyone, we estimate \( p \) using the sample proportion \( \hat{p} \).

\( p \)
True population proportion: fixed but usually unknown.
\( \hat{p} \)
Sample proportion: \( \hat{p} = \frac{x}{n} \) where \( x \) = successes, \( n \) = sample size.

What Is a Confidence Interval?

A confidence interval is a range of plausible values for the population parameter \( p \). It accounts for the variability inherent in sampling: different samples would give different \( \hat{p} \) values, so we build a range instead of relying on a single point estimate.

0.0 0.25 0.50 0.75 1.0
\( \hat{p} = 0.50 \)
0.30 0.70 95% Confidence Interval

Why Do Confidence Intervals Matter?

A single sample proportion \( \hat{p} \) is almost never exactly equal to \( p \). Confidence intervals give us a principled way to quantify uncertainty and justify claims about the population.

Quantify Uncertainty
Wider intervals mean more uncertainty; narrower intervals mean greater precision.
Justify Claims
If a claimed value of \( p \) falls outside the interval, we have evidence against that claim.
Guide Decisions
Policy makers, scientists, and businesses use CIs to make informed decisions under uncertainty.

The Confidence Interval Formula

$$ \hat{p} \;\pm\; z^{*}\sqrt{\frac{\hat{p}(1-\hat{p})}{n}} $$

\( \hat{p} \)
Point estimate: the sample proportion.
\( z^{*} \)
Critical value from the standard normal distribution for the chosen confidence level.
SE
Standard error: \( \sqrt{\frac{\hat{p}(1-\hat{p})}{n}} \) measures sampling variability.

Common Critical Values \( z^{*} \)

\( \hat{p} \)
\( -z^* \)
\( +z^* \)
95%
Confidence Level \( z^{*} \) Interpretation
90%1.645Wider tolerance, narrower interval
95%1.960Most common in AP Statistics
99%2.576Very confident, widest interval

Conditions for Valid Inference

Before constructing a confidence interval for \( p \), you must verify three conditions. Failing to check these is a common AP exam deduction.

1
Random: The data must come from a random sample or randomized experiment.
2
Independence (10% condition): \( n \leq 0.10 \cdot N \), where \( N \) is the population size. This ensures observations are approximately independent.
3
Normal (Large Counts): \( n\hat{p} \geq 10 \) and \( n(1-\hat{p}) \geq 10 \). This ensures the sampling distribution is approximately normal.

Confidence Level vs. Confidence Interval

These two terms sound similar but mean different things. Confusing them is one of the most frequent mistakes on the AP exam.

Feature Confidence Level Confidence Interval
What it isA percentage (e.g., 95%)A range (e.g., (0.42, 0.58))
MeaningLong-run capture rate of the methodSpecific range from one sample
Fixed or varies?Chosen by the researcherVaries from sample to sample
Symbol\( C \) or \( 1 - \alpha \)\( (\hat{p} - E,\;\hat{p} + E) \)

How to Justify a Claim

When the AP exam asks you to "justify a claim based on a confidence interval," follow this structured approach:

1
State the parameter: Define \( p \) in context (e.g., "the true proportion of voters who support the policy").
2
Check conditions: Verify Random, Independence (10%), and Normal (Large Counts).
3
Compute the interval: \( \hat{p} \pm z^{*}\sqrt{\frac{\hat{p}(1-\hat{p})}{n}} \).
4
Interpret and conclude: State whether the claimed value falls inside or outside the interval. If outside, you have evidence against the claim.

Example 1: School Survey

In a random sample of 200 students, 120 said they prefer online homework. Construct a 95% confidence interval for the proportion of all students who prefer online homework.

Example 1: Interpretation

0.40 0.50 0.60 0.70 0.532 0.668 0.60 p = 0.50?

Example 2: Quality Control

A factory claims that no more than 5% of its products are defective. A random sample of 400 items finds 28 defective. Does the confidence interval support the factory's claim at the 95% level?

Example 3: Election Poll

A pollster surveys 1000 randomly selected voters and finds that 540 support Candidate A. A news anchor claims the race is "a statistical tie." Does a 99% confidence interval support this claim?

What Affects Interval Width?

Factor Change Effect on Width
Sample size \( n \)Increase \( n \)Narrower (more precise)
Sample size \( n \)Decrease \( n \)Wider (less precise)
Confidence level \( C \)Increase \( C \)Wider (more confident)
Confidence level \( C \)Decrease \( C \)Narrower (less confident)
\( \hat{p} \) valueCloser to 0.50Wider (max variability)

Common Mistakes

✕ Wrong Interpretation
Saying "there is a 95% probability that \( p \) is in this interval." Once computed, \( p \) is either in it or not. The 95% refers to the method, not this specific interval.
✕ Skipping Conditions
Forgetting to check Random, 10%, or Large Counts. On the AP exam, this costs points even if your calculation is correct.
✕ Confusing \( p \) and \( \hat{p} \)
Using \( p \) (unknown parameter) in the formula instead of \( \hat{p} \) (known sample statistic). The CI formula requires \( \hat{p} \).

Summary: Key Takeaways

1
Formula: \( \hat{p} \pm z^{*}\sqrt{\frac{\hat{p}(1-\hat{p})}{n}} \)
2
Conditions: Random, 10% rule, and Large Counts (\( n\hat{p} \geq 10 \), \( n(1-\hat{p}) \geq 10 \)).
3
Interpretation: "We are C% confident that the true proportion is between [lower, upper]."
4
Justifying a claim: If the claimed value is outside the interval, there is convincing evidence against the claim. If inside, the data are consistent with the claim.

Question 1 of 5

A 95% confidence interval for a population proportion is (0.62, 0.78). A researcher claims that \( p = 0.50 \). Based on the interval, what can you conclude?

Question 2 of 5

Which of the following is a correct interpretation of a 95% confidence interval?

Question 3 of 5

In a random sample of 500 adults, 320 support a new policy. What is the standard error of \( \hat{p} \)?

Question 4 of 5

Which condition for constructing a confidence interval for \( p \) is described by "\( n\hat{p} \geq 10 \) and \( n(1-\hat{p}) \geq 10 \)"?

Question 5 of 5

A researcher wants a 95% confidence interval that is narrower than the one she currently has. Which action would accomplish this without changing the confidence level?