In one line: To show that behaviour has changed, measure it more than once, then rule out the ordinary reasons it might look different.
A claim like "the therapy worked" only means something if the change was measured. That means recording a behaviour at more than one time and comparing.
The catch is that two measurements are never exactly the same. People have good days and bad days. They get used to a test. Scores drift by chance. Good measurement tells a real change apart from this everyday noise.
Remember: Measure twice, then ask: real change, or just noise?
Free preview
This is the free notes preview
You're reading the free notes. Aimnova Pro unlocks the full study experience — and you can try it with your first topic free to keep:
- FlashcardsLock in vocabulary and key terms with spaced repetition.
- Practice questionsAnswer exam-style questions and get instant AI marking.
- Mock exams & past-paper vaultSit full mocks and see exactly how examiners award marks.
- Personalised study planA daily plan built around your exam date and weak areas.
Key idea: Psychologists capture change by measuring the same thing at different times: over minutes, weeks or many years.
Designs that capture change
Before and after
Measure once, do something, measure again.
A memory test before and after a week of sleep training shows whether scores shifted.
Repeated measures
The same people are tested several times, so each person is compared with themselves.
Differences between people cannot explain the result.
Longitudinal
The same people are followed for months or years and measured again and again.
It tracks slow change, such as growing up or growing old.
Before and after · Repeated measures · Longitudinal
Take a school testing a new revision app. It measures test scores in March and again in June (before and after). Each student's June score is compared with their own March score (repeated measures). Following the same students through the whole of next year would be longitudinal.
The quick alternative is to compare different age groups at one moment, which is called a cross-sectional study. It is fast, but the groups are different people, so it shows a difference and not a change.
Watch out: Measuring people twice creates a problem of its own.
People often do better on a test simply because they have seen it before. This is a practice effect.
Practice with exam-style questions
Answer exam-style questions and get AI feedback that shows you exactly what examiners want to see in a full-marks response.
Key idea: Before believing a change, check three things: is the measure consistent, is there a comparison, and is the change big enough?
Three checks
A consistent measure
The same tool is used both times, and it gives steady results.
A wobbly measure invents change that is not there.
A comparison
A second group is measured twice but gets no treatment (a control group).
It shows what would have changed anyway: a holiday, the season, growing older.
Big enough
A tiny shift may just be chance.
A change that is clearly larger than the normal ups and downs is more convincing.
Consistent · Compared · Big enough
Back to the revision app. Scores rose by 20 points. But if the test was a quick quiz that gives different results each time, or every class improved by June anyway, or the rise was really one mark, the "change" cannot be trusted.
Extreme scores drift back: People who are picked because they scored very low usually score a little higher next time, by chance alone.
This is regression to the mean. It can make a useless treatment look as if it worked.
Key idea: A treatment is effective if it produces a real change that lasts, in the people it was meant for.
Three questions about any treatment
Compared with what?
The treated group should improve more than a group that was not treated.
Does it last?
A change measured the week after a course may be gone in six months.
Measuring again later is called a follow-up.
How many people?
For a whole town or country, psychologists count how common a problem is (its prevalence).
If a campaign against smoking works, the prevalence of smoking falls.
Compared with what · How long · How many
So effectiveness can be judged at three levels: for one individual, for a school or a town (local), or across countries (global).
Use it in an answer: When a study claims that something "worked", ask the three questions.
Each one that was not answered is a weakness in the claim.
Feeling unprepared for exams?
Get a clear study plan, practice with real questions, and know exactly where you stand before exam day. No more guessing.
How this is asked: Studies that claim a programme changed something are common in the 15-mark question on a study you are given.
Measurement and causality are the two concepts that fit them best.
The pattern
- Name the design: was it before and after, with or without a control group?
- Check the measure: the same tool, and a steady one?
- Ask what else could explain the change: practice, time, extreme scores.
- Say whether the change was big enough, and whether it lasted.
- Conclude: can the study claim that the programme caused the change?
The trap: "scores fell, so it worked": A fall from before to after has many possible causes.
Only a comparison group shows what the programme added.
Discuss the following study with reference to two or more of the following concepts: bias, causality, measurement and/or responsibility.
A school wanted to know whether a six-week mindfulness programme reduces stress. Thirty students who had told their tutors that they felt very stressed were invited to take part. Each completed a ten-question stress questionnaire in the week before the programme and the same questionnaire in the week after it. The average stress score fell from 32 to 26. The school concluded that mindfulness reduces stress and made the programme compulsory for all students.
Model answer plan
See the answer broken into clear points, with guidance on how each point earns marks.