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

Experimental Method » Independent Measures and Random Allocation

What you'll learn this session

Study time: 30 minutes

Cambridge spec: 7.1.12, 7.1.13

What independent measures means and how it differs from repeated measures
How random allocation works and why researchers use it
The strengths and weaknesses of independent measures
How to choose between the two designs in a new scenario and justify your choice

🔒 Unlock Full Course Content

Unlock This Course

Independent measures: different people in each condition

In the last lesson, Experimental Design and Repeated Measures, the same people took part in every condition. There is one other design you need to know. This time, each person takes part in only one condition.

Key terms:

Independent measures: an experimental design where different participants take part in each condition of an experiment, so each participant experiences only one of the levels of the independent variable (IV).
Random allocation: assigning participants to the different conditions in an experiment so that each participant has an equal chance of being assigned to any particular condition.

Learn this

Independent measures = different participants in each condition. If the IV has two levels, you need two separate groups of people. Each person is measured on the dependent variable (DV) once.

Worked example

A researcher wants to know whether wearing noise-cancelling headphones helps students to finish a word search faster. Forty students take part. Twenty students do the word search wearing headphones. The other twenty do the same word search with no headphones. The researcher records the time taken in seconds.

The IV is the headphones (worn or not worn). The DV is the time taken. Each student does only one condition, so this is independent measures.

Random allocation

With independent measures there is a risk. Suppose the twenty fastest readers all ended up in the headphones group. That group would finish first, but that would be down to the people, not the headphones. The researcher needs a fair way to decide who goes where.

That fair way is random allocation. Every participant has the same chance of ending up in any condition, and the researcher does not choose who goes where.

🎩 Names from a hat

Write each participant's name on a slip of paper. Put the slips in a hat and mix them. Draw out half the slips for the headphones group. The names left in the hat go in the no-headphones group.

🎲 Another fair method

Toss a coin for each person as they arrive. Heads means the headphones group and tails means the no-headphones group. Every person has a 50% chance of each condition.

Because chance decides, the groups are likely to be mixed in a similar way, with some fast readers and some slow readers in each. Any difference in the DV is then more likely to be due to the IV.

Do not mix these up

Random allocation is about which condition a participant is put in, after the participants have been found. It is not the same as how the participants were chosen in the first place. Choosing participants is a separate topic that you will meet later in the course.

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: each person does only one condition, so there are no practice or tiredness effects from doing a second condition. This makes it more likely the IV caused the difference.

Validity: participants see only one condition, so they are less likely to guess what the study is about and change how they behave.

Practical: the same task can be used in each condition, so no second version needs to be made up, and no counterbalancing is needed.

❌ Weaknesses

Validity: the groups contain different people, so differences between the people (such as reading speed) may explain the difference in scores, not the IV. Random allocation helps, but cannot guarantee equal groups.

Reliability: a repeat of the study would use different people, so the groups may be different and the results may not be consistent.

Practical: more participants are needed than for repeated measures, which takes more time and effort.

Notice how the two designs mirror each other. A strength of one is often a weakness of the other.

Choosing between the two designs

In the exam you may be given a new study and asked to identify the design, or to say which design a researcher should use and why. Use these steps.

Ask: are the same people in every condition? If not, it is independent measures.
If asked to choose, look for a problem in the scenario that one design would cause.
Give a reason that uses the scenario's own details.

📝 Clue for independent measures

"One group of pupils learned a poem by reading it aloud. A different group learned it silently." Different people, so independent measures.

🧠 Clue for repeated measures

"Each pupil learned one poem aloud and a second poem silently." Same people in both, so repeated measures.

Worked example: justifying a choice

A researcher tests how long pupils take to solve a puzzle. Once you know the trick, it is easy. Which design should be used?

Independent measures. If the same pupils did the puzzle in both conditions, they would already know the trick the second time, so their times would change because of practice and not because of the IV. Using different pupils in each condition avoids this, and random allocation makes the groups fair.

Repeated measures would be the better choice when the researcher wants to avoid differences between people, such as when only a small group of volunteers is available, or when the measure varies a lot from person to person, such as how fast someone can run.

Applying it to a new scenario

A teacher wants to know whether pupils learn spellings better if they sing them. Thirty pupils have their names put in a hat. Fifteen names are drawn for the singing group. The other fifteen make up the group who say the spellings. After ten minutes, all thirty take a spelling test.

Design: independent measures, because each pupil learns the spellings in only one way.
Random allocation: names were drawn from a hat, so every pupil had an equal chance of being in either group.
Strength: no pupil has practised the test before, so there are no practice effects.
Weakness: by chance, the better spellers may end up in one group.

Common mistakes

Writing "the same participants" when describing independent measures. It is different participants in each condition.
Saying random allocation means "the researcher picks people who seem fair". It is chance, with no choice by the researcher.
Saying random allocation makes the groups exactly equal. It makes it fair, but it cannot promise it.
Naming a strength or weakness without saying what it does to the results in this study.
Giving a reason for a design choice that is not in the scenario. Use the details you are given.

Exam-style question

A researcher investigates whether eating breakfast helps pupils to concentrate. Sixty pupils take part. The names of all the pupils are put in a bag. Thirty names are drawn out and these pupils eat breakfast before the study. The other thirty eat nothing. All sixty pupils then complete the same concentration test and their scores are compared.

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

(b) Describe how random allocation was used in this study. [2 marks]

Model answer

(a) Independent measures (1).

(b) The names of all the pupils were put in a bag and thirty were drawn out for the breakfast group (1), so every pupil had an equal chance of being in either group (1).

(c) The pupils in the two groups are different people (1), so by chance the pupils who concentrate best may all be in one group, and the difference in scores may not be caused by breakfast (1).

Exam tip

For part (b), do not just write "names in a hat". Say what the researcher did in this study and add that everyone had an equal chance. Use the scenario's own words.

Test Your Knowledge