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

Randomized block design

Randomized block design appears in 11 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 Randomized block design is tested

Example 1Harder
A hospital wants to compare recovery times under three post-surgery care protocols. Surgeons at the hospital frequently differ in their surgical techniques, which could affect recovery. To control for surgeon variability, the researchers use each of 10 surgeons with all three protocols, randomly assigning different patients to protocols within each surgeon's patients. This design is best described as:
  1. AA randomized complete block design, where surgeons are the blocks.Correct
  2. BA matched pairs design, because surgeons are paired with treatments.
  3. CA completely randomized design, because patients are randomly assigned to protocols.
  4. DA cluster sample, because surgeons' patients are sampled as a group.

Why A is right

A randomized complete block design uses blocks to account for a known source of variability (here, surgeon-to-surgeon differences). Each surgeon serves as a block, and all three protocols are randomly applied within each surgeon's patients. This removes surgeon variability from the experimental error, increasing the ability to detect real differences between protocols. The 'complete' designation means all three treatments appear within each block (each surgeon's patients).
Why the other 3 choices miss
B
Incorrect. A matched pairs design involves pairing two similar individuals (or using the same individual twice) for comparing two treatments. Here, there are three protocols and 10 blocks — this is a block design, not matched pairs.
C
Incorrect. While patients are randomly assigned to protocols, the design specifically uses surgeon as a blocking factor to account for known variability. Ignoring the blocking structure and calling it a CRD would fail to use the precision advantage of blocking.
D
Incorrect. Cluster sampling is a survey sampling technique where groups are selected as units. Here, the design is experimental with random assignment. Surgeons are experimental blocks, not sampled clusters.

Studied alongside

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