Skip to content
AP Statistics · Inference for Categorical Data: Proportions

Justify claim

Justify claim appears in 6 reviewed AP Statistics questions and sits in Inference for Categorical Data: Proportions. Below are worked examples showing exactly how the exam tests it, with the answer marked and every distractor explained.

How Justify claim is tested

Example 1Easier
A public library wants to know whether the type of reminder affects how quickly overdue books come back. For one month, each borrower with an overdue book was randomly assigned to get either a text-message reminder or an email reminder. Let p_T be the proportion of all such borrowers who would return the book within one week after a text reminder. Let p_E be the same proportion after an email reminder. A 95 percent confidence interval for p_T - p_E, built from the results, is (0.04, 0.15). Which of the following claims is supported by the interval?
  1. ABecause every value in the interval is positive, there is convincing evidence that a greater proportion of text-reminder borrowers return books within a week.Correct
  2. BBecause the interval does not contain 0, there is no convincing evidence of a difference between the two reminder methods.
  3. CBecause 0.04 is so close to 0, the data do not give convincing evidence that text reminders work better.
  4. DBetween 4% and 15% of borrowers who receive a text reminder return their books within a week.

Why A is right

The interval estimates p_T - p_E, and both endpoints (0.04 and 0.15) are greater than 0. So every plausible value of the difference is positive, and 0 (no difference) is not a plausible value at the 95 percent confidence level. That gives convincing evidence that p_T > p_E, meaning a larger proportion of borrowers return their books within a week after a text reminder than after an email reminder. Because the treatments were randomly assigned, the difference can be credited to the type of reminder.
Why the other 3 choices miss
B
Reverses the decision rule: an interval that leaves out 0 is evidence OF a difference, not evidence of no difference.
C
Judges the evidence by how close an endpoint is to 0. The question is whether 0 is inside the interval, and here it is not, however small the lower endpoint is.
D
Reads the interval for the difference p_T - p_E as if it were an interval for the single proportion p_T. The values 0.04 to 0.15 describe how much higher the text-reminder return rate is, not the return rate itself.

Studied alongside

Other concepts from Inference for Categorical Data: Proportions.