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

Replication

Replication appears in 8 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 Replication is tested

Example 1Medium
An experiment uses a 2×3 factorial design with factors A (2 levels) and B (3 levels). Each treatment combination is replicated 5 times. What is the total number of experimental units in this study?
  1. A30Correct
  2. B25
  3. C10
  4. D36

Why A is right

Number of treatment combinations = 2 × 3 = 6. With 5 replicates per treatment combination: total experimental units = 6 × 5 = 30. Each of the 6 cells (A₁B₁, A₁B₂, A₁B₃, A₂B₁, A₂B₂, A₂B₃) has 5 subjects, giving 30 total.
Why the other 3 choices miss
B
Incorrect. 25 does not arise from standard calculation of this design. It might result from (2+3) × 5 = 25, which adds factor levels instead of multiplying them, giving an incorrect number of treatment combinations.
C
Incorrect. 10 = 2 × 5 accounts for only one factor (A) with 5 replicates each, ignoring the 3 levels of factor B. All 6 treatment combinations must be counted.
D
Incorrect. 36 = 6 × 6, which would result if there were 6 replicates instead of 5, or from a 2×3 design with 6 replicates per cell. With 5 replicates: 6 × 5 = 30.
Example 2Easier
The purpose of replication in an experiment is to:
  1. ABlind the subjects to their treatment assignment
  2. BEnsure the experiment can be repeated by other researchers
  3. CUse enough experimental units in each treatment group to reduce the effects of chance variation and increase the ability to detect a real treatment effectCorrect
  4. DMatch subjects in pairs before random assignment

Why C is right

Replication — having multiple experimental units per treatment — reduces the impact of individual variability on the results. With more units per group, the sample means become more stable, and the experiment has greater power to detect real treatment differences.
Why the other 3 choices miss
A
Incorrect because blinding is a separate design principle related to preventing bias, not to sample size within treatment groups.
B
Incorrect because while reproducibility (other researchers repeating the study) is important in science, "replication" in experimental design specifically refers to having multiple units per treatment.
D
Incorrect because matching subjects in pairs is the matched pairs design principle, not replication.

Studied alongside

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