AP Statistics Unit 6: Inference for Categorical Data: Proportions
Confidence intervals and significance tests for proportions.
Unit overview
Inference for proportions answers two kinds of question. A confidence interval estimates an unknown proportion: . A significance test asks whether data give convincing evidence against a claimed value , using .
Each procedure has the same four parts: state the parameter and hypotheses or confidence level, check the random, 10% and large counts conditions, do the calculation, and conclude in context. The same structure extends to a difference of two proportions, where the test uses a pooled proportion.
Key topics
- Confidence intervals for
- Point estimate plus or minus margin of error; interpreting the interval and the confidence level.
- Significance tests for
- Hypotheses, test statistic, P-value and conclusion at a significance level.
- Errors and power
- Type I error rejects a true ; Type II fails to reject a false one; power is .
- Two-proportion procedures
- Intervals and tests for , with pooling in the test.
Worked example
In a random sample of 400 voters, 220 support a measure. Find the 95% confidence interval's margin of error (use ).
- .
- .
- .
Answer:
Common mistakes
- Writing hypotheses about instead of the parameter .
- Saying there is a 95% probability that the parameter is in this particular interval; the 95% describes the method.
- Concluding that is true when the P-value is large; the correct phrase is that there is not convincing evidence for .
Exam tips
- Use in the standard error for a test and for an interval.
- Link the conclusion to the P-value and explicitly, and keep it in the language of the problem.
More practice on this site
- Descriptive Statistics Study Guide (study guide)
- Statistics Formulas cheat sheet (one-page formulas)
- Descriptive Statistics Quiz (quick quiz)
- Statistics practice problems (with worked solutions)
See the AP Statistics exam format and every unit on the course page.