A within-subjects design is an experimental design in which the same group of participants is exposed to all independent variable levels. This design controls for individual differences and often requires fewer participants.
A within-subjects design allows researchers to assign test participants to different treatment groups.
In a within-subjects design, each participant experiences every condition of the independent variable.
Key Takeaways
- Definition: A within-subjects design (also called a repeated-measures or within-groups design) tests the same participants in every condition of the independent variable.
- Fewer Participants: Because each person provides data for every condition, within-subjects designs need smaller samples than between-subjects designs of the same power.
- No Participant Variables: Individual differences such as IQ or mood are controlled automatically, since each participant acts as their own control.
- Order Effects: Repeating conditions risks practice, fatigue, and carryover effects that between-subjects designs avoid.
- Counterbalancing: Varying the order conditions are presented in spreads order effects evenly, using schemes like ABBA or Latin squares.
- Statistical Test: A within-subjects design needs a related-samples test, such as the related t-test or Wilcoxon signed-ranks test.
How it Works
Researchers test the same participants repeatedly across every treatment condition, comparing scores within each person rather than between separate groups.
There is no separate control group. Instead, each participant is tested before and after exposure to the treatment.
This is why the approach is called “within-subjects”: the comparison happens within the same group of people. A between-subjects design works the opposite way, comparing different groups of participants instead.
Using a within-subjects design
Within-subjects studies are typically used for longitudinal studies, as researchers can assess changes within the same group of subjects over an extended period of time.
Consider a psychiatrist testing medication for Obsessive-Compulsive Disorder (OCD). She has four drug doses to compare.
To determine which dose works best, she measures each patient’s performance four times, once after each of the four doses for a week.
Each patient’s performance is thus measured at each of the four dose levels.
Advantages
Does not require a large subject pool
Within-subjects designs require smaller sample sizes as each participant provides repeated measures for each treatment condition.
This also reduces the cost and resources necessary to conduct these studies.
On the other hand, a between-subjects study would require at least twice as many participants as a within-subject design. This also means twice the cost and resources.
No variation in individual differences
Within-subjects designs control for individual differences because the same participants complete every condition. Each participant acts as their own control.
Stable traits such as IQ, personality, baseline mood, and prior experience stay constant across conditions. This shrinks the error variance a test compares the treatment effect against, so a smaller true effect becomes detectable.
Fewer participants are needed for the same statistical power, because each person contributes data to every condition rather than just one.
Disadvantages
Time
Data collection can take a long time since each participant is given multiple treatments. In addition, it can be challenging to control the effects of time on the study’s outcomes.
Between-subjects studies tend to have shorter sessions than within-subject ones.
Order effects
Because participants complete every condition, performance can shift simply from repeating, tiring of, or carrying over from an earlier condition, not from the independent variable itself. This is called an order effect.
It takes three distinct forms, each with a different fix.
- Practice effects: participants get better with repetition as they grow familiar with the task, apparatus, or timing.
- Fatigue and boredom effects: participants tire or lose motivation, so scores in later conditions fall.
- Carryover effects: a treatment’s effect lingers into the next condition, such as a drug not yet cleared from the body.
For example, exposure to a reaction time test could make participants’ reaction times faster in a subsequent treatment due to familiarity with the study.
Carryover is the hardest of the three to fix, because it does not always affect every condition equally.
Counterbalancing: reducing order effects
Counterbalancing is the systematic variation of the order in which conditions are presented, so that any order effect is spread evenly across conditions rather than piling up in one.
It is the standard remedy for order effects in a within-subjects design.
- ABBA counterbalancing: each participant receives the sequence A → B → B → A (or B → A → A → B), which cancels out a steady practice effect.
- Between-subject counterbalancing: half the participants get order A-then-B, and the rest get B-then-A, spreading order effects across the sample.
- Latin square designs: with more than two conditions, each condition appears once in each position across a set of orders, balancing all conditions across the sequence.
Counterbalancing works well for practice and fatigue, which build up gradually.
It does not fix carryover that is uneven between conditions. When that risk is high, researchers space conditions apart with a washout period, or switch to a between-subjects design instead.
Choosing the right statistical test
Because the same participants provide every score, a within-subjects design needs a related-samples statistical test rather than the tests used for independent groups.
- Nominal data: the sign test.
- Ordinal data: the Wilcoxon signed-ranks test.
- Interval or ratio data: the related (paired) t-test, or repeated-measures ANOVA for three or more conditions.
Picking the wrong test family is a common student error.
A related-samples test assumes the scores in each condition come from the same people, which a within-subjects design guarantees.
Examples
- A course of Cognitive Decline in Hematopoietic Stem Cell Transplantation: A Within-subjects Design (Friedman et al., 2009).
- Beliefs, attitudes, and intentions toward nuclear energy before and after Chernobyl in a longitudinal within-subjects design (Verplanken, 1989).
- A comparison of paper and online tests using a within-subjects design and propensity score matching study (Lottridge, Nicewander, & Mitzel, 2011).
- A test of exercise analgesia using signal detection theory and a within-subjects design (Fuller and Robinson, 1993).
- Reported jealousy differs as a function of menstrual cycle stage and contraceptive pill use: A within-subjects investigation (Cobey et al., 2012).
- Behavioral effects of haloperidol in young autistic children: An objective analysis using a within-subjects reversal design (Cohen et al., 1980).
- Tipping and service quality: A within-subjects analysis (Lynn and Sturman, 2010).
Learning Check
Which of the following are advantages of using a within-participant design in experimental research?
- It controls for individual differences, as the same participants are used in all conditions.
- It requires fewer participants compared to between-participants designs.
- It allows for observing changes over time or conditions in the same individual.
- All of the above.
- It is unaffected by order or sequence effects.
- It allows for the study of interactions between independent variables.
- It requires less time to collect data because each participant is in all conditions.
- None of the above.
Answers
4. Note: Option 5 is incorrect because within-participant designs can indeed be affected by order or sequence effects. Option 7 can also be incorrect because, even though each participant is in all conditions, each condition may require separate sessions, potentially lengthening the data collection time.
Option 6 is not exclusively an advantage of within-participant designs and can apply to between-participants designs as well.
Frequently Asked Questions
1. What’s the difference between a between-subjects versus within-subjects design?
Between-subjects and within-subjects designs are two different methods for researchers to assign test participants to different treatments.
In a between-subjects design, researchers will assign each subject to only one treatment condition. In contrast, in a within-subjects design, researchers will test the same participants repeatedly across all conditions.
Between-subjects and within-subjects designs can be used in place of each other or in conjunction with each other.
Each type of experimental design has its own advantages and disadvantages, and it is usually up to the researchers to determine which method will be more beneficial for their study.
2. Can you use a between-subjects and within-subjects design in the same study?
Yes. Between-subjects and within-subjects designs can be combined in a single study when you have two or more independent variables (a factorial design).
Factorial designs are a type of experiment where multiple independent variables are tested. Each level of one independent variable (a factor) is combined with each level of every other independent variable to produce different conditions.
Each combination becomes a condition in the experiment. In a factorial experiment, the researcher has to decide for each independent variable whether to use a between-subjects design or a within-subjects design.
In a mixed factorial design, researchers will manipulate one independent variable between subjects and another within subjects.
References
Allen, M. (2017). The sage encyclopedia of communication research methods (Vols. 1-4). Thousand Oaks, CA: SAGE Publications, Inc doi: 10.4135/9781483381411
Baeyens, F., Díaz, E., & Ruiz, G. (2005). Resistance to extinction of human evaluative conditioning using a between‐subjects design. Cognition & Emotion, 19(2), 245-268.
Birnbaum, M. H. (1999). How to show that 9> 221: Collect judgments in a between-subjects design. Psychological Methods, 4(3), 243.
Carey, A. A., Lester, T. G., & Valencia, R. M. (2016). The Effects of a Fatal Vision Goggles Intervention on Middle School Aged Children’s Attitudes toward Drinking and Driving and Texting and Driving as Related to Impulsivity: A Between Subjects Design (Doctoral dissertation, Brenau University).
Chang, H. I., & Kang, W. C. (2018). The Impact of the 2018 North Korea-United States Summit on South Koreans’ Altruism Toward and Trust in North Korean Refugees: Between-Subjects Design Around the Summit. Available at SSRN 3270334.
Egele, V. S., Kiefer, L. H., & Stark, R. (2021). Faking self-reports of health behavior: a comparison between a within-and a between-subjects design. Health psychology and behavioral medicine, 9(1), 895-916.
Ehrlichman, H., Brown Kuhl, S., Zhu, J., & WRRENBURG, S. (1997). Startle reflex modulation by pleasant and unpleasant odors in a between‐subjects design. Psychophysiology, 34(6), 726-729.
Jhangiani, R. S., Chiang, I.-C. A., Cuttler, C., & Leighton, D. C. (2019, August 1). Experimental Design. Research Methods in Psychology. Retrieved from https://kpu.pressbooks.pub/psychmethods4e/
