Test Your Knowledge Flash Cards 🔒Play Lemonaire 🔒Play Last Stand

Experimental Method » Experimental Design and Repeated Measures

What you'll learn this session

Study time: 30 minutes

Cambridge spec: 7.1.8, 7.1.9, 7.1.10, 7.1.11

What an experimental design is, and the two designs you need to know
How repeated measures works and how counterbalancing helps
The strengths and weaknesses of repeated measures for validity and reliability
How to spot and apply repeated measures in a new scenario

🔒 Unlock Full Course Content

Unlock This Course

What is an experimental design?

Before an experiment starts, the researcher has to decide who does what. If the independent variable (IV) has two levels, such as a quiet room and a noisy room, do the same people try both rooms, or do different people try each one? That decision is the experimental design.

Key terms:

Experimental design: a way of assigning participants to different conditions or levels in an experiment.
Repeated measures: an experimental design where the same participants take part in each condition of an experiment, so they experience all of the levels of the independent variable.
Counterbalancing: changing the order in which participants experience the different conditions, to reduce the impact of extraneous variables in an experiment.

You need to know two designs: repeated measures (this lesson) and independent measures (the next lesson, Independent Measures and Random Allocation). You do not need any other designs.

How repeated measures works

In repeated measures, every participant does every condition. Each person is measured on the dependent variable (DV) more than once, so each person is compared with themselves.

Learn this

Repeated measures = the same participants in every condition. If the IV has two levels, each participant takes part twice, once at each level.

Worked example

A researcher asks whether the smell of peppermint helps students solve anagrams. Twenty students solve one set of ten anagrams in a room with a peppermint smell. The same twenty students then solve a different set of ten anagrams in a room with no smell. The IV is the smell (peppermint or none). The DV is the number of anagrams solved correctly. Every student experiences both levels of the IV, so this is repeated measures.

Spotting the design in a new scenario

In the exam you will be given a short study and asked to identify the design. Ask one question: are the same people in every condition?

Yes, the same people did both conditions: repeated measures.
No, different people did each condition: independent measures.

🔄 Repeated measures

"Each of the 15 cyclists rode the same track once on a flat bike and once on a bike with a heavy backpack. Lap times were compared."

The same cyclists did both conditions.

👥 Not repeated measures

"One class of 30 pupils played a quiz in teams. A different class of 30 pupils played alone. Scores were compared."

Different people in each condition, so this is the other design.

Watch for clue words such as "the same participants", "each person did both" or "before and after the break". If the question asks you to explain or justify, give a reason from the scenario. If it only says identify, the name is enough.

The problem: order effects

Because people do the conditions one after another, the order can change their scores. This brings in extraneous variables (variables other than the IV that could affect the DV, which you will study properly later in Extraneous Variables and Controls). Two common ones are:

Practice: in the second condition people are more used to the task, so they may do better whatever the IV is.
Tiredness (or boredom): in the second condition people may be worn out or fed up, so they may do worse whatever the IV is.

Suppose all the students in the anagram study did the peppermint room first. Their second score might be higher just because they had warmed up, and the researcher might wrongly think no-smell is better. The result would be muddled.

The fix: counterbalancing

Counterbalancing means changing the order in which participants experience the conditions. Half of the participants do condition A first, then condition B. The other half do B first, then A.

📝 Group 1

Peppermint room first, then no-smell room.

📝 Group 2

No-smell room first, then peppermint room.

Now practice helps both conditions equally, and so does tiredness. The effect of order is spread across the conditions instead of always falling on the same one. Counterbalancing reduces the impact of these extraneous variables, but it does not remove the problem completely.

Strengths and weaknesses

Examiners link these points to validity (is the study really testing what it aimed to test?) and reliability (are the procedure and results consistent?).

✅ Strengths

Validity: the same people are in both conditions, so differences between people (such as being good at anagrams) cannot explain the difference in scores. This makes it more likely the IV caused the change.

Reliability: with the same people each time, the comparison is less affected by who happened to be in each group, so the results are more likely to be consistent if the study is repeated.

Practical: only half as many participants are needed as in a design with different people in each condition.

❌ Weaknesses

Validity: order effects such as practice or tiredness may change the DV, so the result may not be due to the IV alone. Counterbalancing helps but may not remove them.

Validity: participants see every condition, so they may work out what the study is about and change how they behave.

Reliability: if the order is not counterbalanced, a repeat of the study with a different order could give different results.

Common mistakes

Writing "different participants" when describing repeated measures. It is the same participants every time.
Saying counterbalancing means "mixing up the participants". It changes the order of the conditions.
Saying counterbalancing gets rid of practice effects completely. It only reduces their impact.
Naming the design with no reason when the question asks you to explain or justify it.
Saying a strength is just "it is easy". Link it to validity, reliability or needing fewer participants.

Applying it to a new scenario

A coach wants to know whether a ten-minute warm-up changes how far athletes can jump. Twelve athletes jump with no warm-up on Monday and after the warm-up on Tuesday. The jump distance in centimetres is recorded.

Design: repeated measures, because the same twelve athletes did both conditions.
Problem: tiredness or practice from Monday could affect Tuesday's jumps.
Counterbalancing: six athletes do no warm-up first and the other six do the warm-up first.
Strength: the athletes are the same both days, so one person being naturally stronger cannot explain the difference.

Exam-style question

A researcher investigates whether a bright classroom helps pupils to read quickly. Twenty pupils read one short story in a bright classroom and then a different short story of the same length in a dim classroom. The number of words each pupil reads in one minute is recorded.

(a) Identify the experimental design used. [1 mark]

(b) Explain one weakness of this design for this study. [2 marks]

Model answer

(a) Repeated measures (1).

(b) Pupils read in the bright room first, so they may be better in the second room because of practice (1) or worse because of tiredness (1), so the difference in reading speed may not be caused by the lighting alone.

Exam tip

For a weakness question, name the problem (such as practice or tiredness), then say what it does to the results in this study. Use the scenario's own words.

Test Your Knowledge