Skip to content
AP Statistics · Exploring One-Variable Data and Collecting Data

Completely randomized design

Completely randomized design appears in 5 reviewed AP Statistics questions and sits in Exploring One-Variable Data and Collecting Data. Below are worked examples showing exactly how the exam tests it, with the answer marked and every distractor explained.

How Completely randomized design is tested

Example 1Easier
A company wants to test four different website layouts (A, B, C, D) to see which generates the most sales. They randomly assign 10,000 visitors to one of the four layouts and track purchases over one week. This is an example of:
  1. AA matched pairs design comparing layouts pairwise
  2. BAn observational study because the company only observed purchase behavior
  3. CA completely randomized experiment with four treatmentsCorrect
  4. DA stratified sampling study of web visitors

Why C is right

This is a completely randomized experiment. The treatment (website layout) is randomly assigned to subjects (visitors), and the response (purchase behavior) is measured. Random assignment allows the company to make causal claims about which layout generates more sales.
Why the other 3 choices miss
A
Incorrect because matched pairs would involve pairing subjects and comparing two conditions; here, there are four independent groups.
B
Incorrect because the company actively imposes a treatment (assigns visitors to specific layouts), making this an experiment.
D
Incorrect because this is an experiment, not a sampling study; visitors are being assigned treatments, not just surveyed.

Studied alongside

Other concepts from Exploring One-Variable Data and Collecting Data.