Repeated measures design
An experimental design in which the same participants take part in every condition, so each person's scores in one condition are compared with their own scores in another.
- In a repeated measures design the same participants take part in every condition.
- Each person acts as their own comparison, because you set their score in one condition against their own score in another.
- Any difference you find cannot be blamed on one group happening to have better memories or faster reactions.
- Same people, every condition.
- Because it is the same people, differences between participants cannot explain the results.
Strengths and Weaknesses
Order effects
Changes in participants' performance caused by the order in which they do the conditions, such as improving through practice or getting worse through fatigue or boredom.
- Strengths
- Participant variables are removed, because the very same people do each condition.
- You need fewer participants than a design that uses a separate group per condition.
- Weaknesses
- Order effects can creep in, as people practise or grow tired between conditions.
- Doing both conditions can let people guess the aim and change how they behave.
- You give the same Year 10 class a memory test in the morning and a similar one in the afternoon.
- The afternoon scores are higher, but that could be practice from the morning rather than the time of day.
- That practice, and not your independent variable, may be what lifted the score.
Counterbalancing
Counterbalancing
A technique for controlling order effects in a repeated measures design, where the order in which participants do the conditions is varied so that half do them one way round and half the other.
- Counterbalancing varies the order in which people meet the conditions, so practice and fatigue do not always fall on the same one.
- In the simplest version, half the class do condition A then B, while the other half do B then A.
- This way the practice and fatigue are spread evenly, so they cancel out across the results.
- Do not claim counterbalancing removes order effects: it balances them across the group, it does not erase them for each person.
- Do not confuse counterbalancing, which varies the order of conditions, with randomising the order of test materials.
Using Repeated Measures Well
- Choose it when the same people can safely do every condition without one spoiling the next.
- When order effects are likely, add counterbalancing: half do A then B, half B then A, so practice and fatigue balance out rather than vanish.
- Keep everything but the IV constant across the conditions, so only the IV could change the scores.
- What is a repeated measures design?
- Give one strength and one weakness of repeated measures.
- What is an order effect, and name two kinds.
- How does counterbalancing deal with order effects?