Experimental Design Statistics Example 1

Follow the full solution, then compare it with the other examples linked below.

Example 1

easy
A farmer wants to test whether a new fertiliser improves crop yield. She applies the new fertiliser to Field A and uses the old fertiliser on Field B. She finds Field A produces more. Can she conclude the new fertiliser is better? Identify the flaw in the experiment.

Solution

  1. 1
    Step 1: The two fields may differ in soil quality, sunlight, water drainage, or other factors that affect yield. These are confounding variables.
  2. 2
    Step 2: The experiment has no randomisation โ€” the fields were not randomly assigned. The farmer compared two different fields, not two equivalent groups.
  3. 3
    Step 3: A better design would split a single field into random plots, randomly assigning the new fertiliser to half the plots and the old fertiliser to the other half.

Answer

No, she cannot conclude the new fertiliser is better. The two fields may differ in soil quality and other factors (confounding variables). Random assignment within the same field would be a better design.
Good experimental design requires controlling for confounding variables. Using two entire fields introduces many uncontrolled differences. Random assignment of treatments to equivalent units helps ensure that any observed differences are due to the treatment rather than pre-existing conditions.

About Experimental Design

Experimental design is the careful planning of experiments to establish cause-and-effect relationships by controlling variables, using comparison groups, and randomly assigning subjects to treatment and control conditions to isolate the effect of interest.

Learn more about Experimental Design โ†’

More Experimental Design Examples