Between-subjects vs Within-subjects

Different people vs the same people

A fundamental distinction in experimental design concerns whether participants experience only one condition or all conditions. In a between-subjects design each participant is exposed to just one condition; in a within-subjects design the same participant experiences every condition in sequence. Within-subjects designs offer greater statistical power and require fewer participants, but introduce threats such as order and carryover effects. Between-subjects designs avoid these threats at the cost of requiring larger samples.

Core Concept

In a between-subjects design, participants are randomly assigned to different conditions and each person experiences only one condition. This approach is also known as an independent-groups design. In a within-subjects design, the same participants are exposed to multiple conditions or measurement occasions, which is why it is also called a repeated-measures design. The choice between the two depends on the research question, practical constraints, and the threats a researcher needs to control.

How It Works: Advantages and Trade-offs

Within-subjects designs offer a marked statistical power advantage: because each participant serves as their own control, individual differences in performance are removed from the error term, making it easier to detect real effects with smaller samples. However, participants' responses to a condition can be influenced by the order in which conditions occur — through fatigue, practice (learning effects), or contamination from a prior condition (carryover effects). Between-subjects designs are immune to these threats because each condition is administered to a different set of participants, but this requires larger overall samples to achieve equivalent power.

Concrete Example and Counterbalancing

Consider a workplace study asking whether a quiet environment or background music improves problem-solving performance. In the between-subjects approach, one group works in a quiet room and another in a room with music. In the within-subjects approach, every participant experiences both conditions. To control order effects, counterbalancing is applied: half the participants encounter the quiet condition first, the other half the music condition first. This arrangement distributes order effects systematically across the experiment, preserving interpretability of the results.

Common Pitfalls and Best Practices

The most common mistake is choosing a design based solely on sample size convenience. A within-subjects design may seem attractive when participants are scarce, but the researcher must first ask whether conditions can contaminate each other — if unavoidable carryover is present, a between-subjects design may be necessary despite its cost. Another pitfall is using a within-subjects design without counterbalancing, which introduces a serious internal validity threat. When studies involve multiple independent variables, mixed designs — in which some factors are between-subjects and others are within-subjects — should also be considered as a flexible alternative.

Key terms

Between-subjects Design: A design in which each participant is assigned to only one experimental condition.
Within-subjects Design: A design in which the same participants experience all conditions; also called repeated-measures.
Carryover Effect: When exposure to one condition influences a participant's response in a subsequent condition.
Counterbalancing: Systematically varying the order of conditions across participants to control order effects.
Mixed Design: A design combining between-subjects factors for some variables and within-subjects factors for others.