Which study design reduces inter-subject variability by measuring the same subjects under different conditions?

Which study design reduces inter-subject variability by measuring the same subjects under different conditions?

• A. Between-subject design.
• B. Within-subject (repeated measures) design.
• C. Cross-sectional design.
• D. Parallel-group design.

Answer: (B)

Measuring the same individuals under different conditions controls for differences between people, so inter-subject variability is reduced. In a within-subject or repeated measures design, each participant experiences every condition, meaning comparisons are made within the same person. This cancels out much of the variability that comes from who the participant is, increasing the study’s power to detect the effect of the condition itself. For example, testing a task with and without caffeine using the same group of participants means any differences due to individual ability are less likely to confound the results. In contrast, using different participants for each condition (between-subject design) leaves inter-subject differences contributing to variability, making it harder to detect the condition effect. A cross-sectional design looks at different people at a single point in time and doesn’t involve repeated measurements across conditions for the same individual. A parallel-group design is another way of saying between-subject groups, not within-subject measurements.

Measuring the same individuals under different conditions controls for differences between people, so inter-subject variability is reduced. In a within-subject or repeated measures design, each participant experiences every condition, meaning comparisons are made within the same person. This cancels out much of the variability that comes from who the participant is, increasing the study’s power to detect the effect of the condition itself. For example, testing a task with and without caffeine using the same group of participants means any differences due to individual ability are less likely to confound the results.

In contrast, using different participants for each condition (between-subject design) leaves inter-subject differences contributing to variability, making it harder to detect the condition effect. A cross-sectional design looks at different people at a single point in time and doesn’t involve repeated measurements across conditions for the same individual. A parallel-group design is another way of saying between-subject groups, not within-subject measurements.