Section 1.6

The Design of Experiments

Observational studies can only show association. To prove standard causation, we must intervene. This is the art of the Experiment.

1

The Language of Experiments

Experimental Unit

The person, object, or well-defined item to which a treatment is applied.

Called a Subject if human.

Factor

A variable whose effect on the response variable is of interest.

LogicLens: Factor vs. Treatment

Students often confuse the variable with the outcome.

  • Factor:The broad category (e.g., "Dosage").
  • Level:Specific values (e.g., 0mg, 50mg, 100mg).
  • Treatment:The specific combination applied (e.g., "50mg Dosage").
2

Control & Blinding

Control Group Baseline

A group that receives no treatment (or a standard treatment). It serves as a baseline to compare against.

Placebo Effect

The phenomenon where subjects show improvement simply because they *believe* they are receiving treatment.

Blinding (The Solution)

Single-Blind
Subject doesn't know what they got. Researcher does.
Double-Blind (Gold Standard)
Neither Subject nor Researcher knows.
3

Design Frameworks

Completely Randomized Design

Experimental units are assigned to treatments purely by chance. No grouping. Simplest method, but high variability.

Randomized Block Design

Group (block) similar individuals together first (e.g., Men / Women), THEN randomize treatments within each block.

"Block what you can, Randomize what you can't."

Matched-Pairs Design

A special case of blocking where the "block" size is exactly 2.
Type A: Two similar subjects (Twins).
Type B: One subject does both treatments (Before/After).

Adaptive Engine

LogicLens Practice Suite

Log in to Access Adaptive Practice

Our AI engine generates unique practice problems based on your progress.