This article covers 3.1.4 Research methods of the AQA GCSE Psychology specification.
What Makes a Study an Experiment
- An experiment investigates the effect of one variable on another under controlled conditions.
- You, the researcher, change the IV and measure the effect on the DV.
- Picture testing whether caffeine speeds up reaction time: you give one group a caffeinated drink and another a decaf drink, then time their responses.
Key Idea
- The defining move is that the researcher manipulates the IV on purpose.
- Controlling the other variables is what lets an experiment show cause and effect.
The Defining Features
- A manipulated independent variable
- The independent variable is the factor you change, and you set up the different conditions yourself, such as caffeine versus decaf.
- A measured dependent variable
- The outcome is recorded, usually as a number, such as reaction time in milliseconds.
- Controlled conditions
- Other variables, such as room temperature and the amount of drink given, are kept the same so they cannot affect the results.
- A test of cause and effect
- The design lets you say whether the IV caused the change in the DV.
Note
- Experiments mostly produce quantitative data, such as scores or times.
- They can also gather some qualitative data, such as what participants said afterwards.
Recognising an Experiment
- Check that the researcher deliberately changed something: that is the IV.
- Check that something was measured as a result: that is the DV.
- Check that other variables were controlled, so any change in the DV can be put down to the IV.
Common Mistake
- If nothing was manipulated, the study is not an experiment.
- Simply measuring or observing behaviour, with no IV, is a different method.
Experimental Design
- Experimental design is the way you decide which participants take part in which condition of your study.
- It settles whether the same person meets every level of the IV, or different people meet different levels.
- That single choice shapes who you recruit, how you compare their scores and which problems you must guard against.
Key Idea
- The design is about who does the conditions, not how many conditions there are.
- The very same study can be run with any of the three designs.
The Three Designs You Can Choose From
- Independent groups (also called independent measures)
- Different people take part in each condition, so nobody is tested twice.
- Repeated measures
- The same people take part in every condition, so each person is compared with themselves.
- Matched pairs
- Different people take part, but each one is paired with a similar partner who does the other condition.
Example
- Say you want to test whether a Year 11 class recalls more words revised with music or in silence.
- Independent groups: one tutor group revises with music, a different tutor group revises in silence.
- Repeated measures: the same pupils revise one word list with music and a second list in silence.
- Matched pairs: you pair pupils by their last mock score, then send one of each pair to music and the other to silence.
Spotting the Design from a Description
- If different people sit in each condition and nobody is paired up, it is independent groups.
- If the same people do every condition, it is repeated measures.
- If different people are paired on a key feature first and then split across conditions, it is matched pairs.
Common Mistake
- Do not confuse matched pairs with repeated measures: matched pairs uses different people, just carefully paired.
- Do not judge the design by counting conditions, because any of the three can have two or more.
Strengths and Weaknesses of Each Design
- Independent groups
- Strength: there are no order effects, because each person does only one condition. Weakness: participant variables can differ between the two groups, and you need more people.
- Repeated measures
- Strength: participant variables are removed, because the same people do every condition, and you need fewer people. Weakness: order effects such as practice or fatigue, which counterbalancing helps to control.
- Matched pairs
- Strength: it reduces participant variables by pairing people on key features, with no order effects. Weakness: matching takes time and you can never match on every variable.
- The core trade-off: repeated measures cancels participant variables but risks order effects, while independent groups avoids order effects but risks participant variables.
Self review
- In one sentence, what is an experiment?
- Name the three experimental designs.
- Give one strength and one weakness of repeated measures.
- Give one strength and one weakness of independent groups.
- How does matched pairs reduce participant variables?