High School

AP Statistics exam scripts

academics AP exams statistics & probability

By Brian B.

On the AP Statistics Exam, being precise with your wording can make a huge difference in your Free Response score. Below are templates to help you phrase your explanations of key statistical concepts. These phrasings help ensure you have covered all important aspects of the analysis in a clear and concise manner. If you memorize them, and know how to use them on exam day, you’ll have a terrific advantage!

Terms in [brackets] should be replaced by the relevant number or variable, stated in the context of the problem (including the proper units, when relevant!).

Confidence intervals

It can be stated with ___% confidence that the population [mean, proportion, etc.] of _(context)_ is between [lower bound] and [upper bound].

Confidence levels

If all possible samples of size n were taken, and a 95% confidence interval was generated for each one, 95% of those intervals would contain the population [mean, proportion, etc., in context]; the other 5% would miss.

NOTE: Be sure to adjust this if a different level of confidence (e.g. 99%) is given!

Interpreting p-values

Assuming H₀ and the probability model are true, the probability of getting a test statistic as or more extreme as the one obtained in this sample is [p].

NOTE: “As or more extreme” is best for two-tailed tests. For one-tailed tests, indicate the direction (“as high or higher” or “as low or lower”)

Analyzing or drawing conclusions about p-values

Because p < [α], H₀ is rejected. There is statistically significant evidence to suggest _(H_A, in context)_.

Because p > [α], the test FAILS to reject H₀. There is NOT statistically significant evidence to suggest _(H_A, in context)_.

Slope of a linear regression

[y] is predicted to [increase/decrease] by [slope] [units] for each 1 [unit] increase in [x].

Y-Intercept of a linear regression

When [x] = 0, [y] is predicted to be [y-intercept] [units].

Interpreting r²

[r2]% of the variation in [y] can be accounted for (or explained by) its linear relationship with [x].

NOTE: Make sure you convert the decimal to a percentage, e.g. 0.78 à 78%.

Brian B.

Brian studied History and Political Science at Villanova University, where he graduated magna cum laude and Phi Beta Kappa. After joining Teach for America and working as a teacher for eight years, he decided to pursue his JD at Harvard Law School.

Work with Brian B. Meet all our coaches

Comments