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

Variability reduction

Variability reduction appears in 7 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 Variability reduction is tested

Example 1Medium
A researcher studying the effect of music on productivity randomly assigns 60 office workers to one of three conditions: silence, classical music, or pop music. Workers complete a set of tasks, and their output is measured. The researcher blocks by department before randomizing. The purpose of blocking by department is to
  1. Aeliminate the need for a separate control group in the experiment
  2. Bincrease the total sample size of the experiment to improve its power
  3. Callow each department to choose its preferred music type
  4. Dreduce variability from department-level differencesCorrect

Why D is right

Blocking by department ensures that each department is represented in all three treatment groups. This reduces the variability within treatment groups that might be caused by department-level differences (different job types, work cultures, baseline productivity levels). By controlling for this source of variation, blocking increases the ability to detect treatment effects.
Why the other 3 choices miss
A
Incorrect. Blocking does not eliminate the need for a control group. The silence condition serves as the control, and blocking is an additional design feature that works alongside control and randomization.
B
Incorrect. Blocking does not change the total number of participants. It reorganizes how participants are assigned to treatment groups.
C
Incorrect. Blocking does not allow participants to choose their treatment. Within each department, workers are randomly assigned to conditions. Blocking structures the randomization, not the choice.

Studied alongside

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