This article covers 3.1.4 Research methods of the AQA GCSE Psychology specification.
Planning for Reliability and Validity
- Reliability is the consistency of your results, while validity is whether you are measuring the thing you actually meant to.
- You plan for both from the very start, because you cannot repair them once the data is in.
- Almost every design choice you make touches one or both of them.
Key Idea
- Reliability and validity are not the same target: a study can be highly consistent yet still measure the wrong thing.
- For each choice, decide whether it protects consistency, accuracy or both.
Standardisation
- Keep the instructions, timings and conditions identical for every participant.
- A repeat of the study should then run the same way, which supports reliability.
- It also removes stray differences between conditions that could otherwise explain the results.
Operationalised Measures
- Define each variable by exactly how you will measure it, such as memory as the number of words recalled in two minutes.
- A precise measure can be applied the same way each time, which lifts reliability.
- It also pins the measure to the concept you really care about, which supports validity.
Pilot Studies
Definition
Pilot study
A small trial run of a study, done before the real thing, to check the procedure and materials work and to fix any problems in advance.
- Run a small trial first to test the instructions, materials and timings.
- You catch confusing wording, tasks that are too easy or too hard, and hidden extraneous variables.
- Fixing these before the real study protects both reliability and validity.
Appropriate Sampling
- Choose a sampling method that makes the sample represent the target population.
- A representative sample supports population validity, so the findings generalise beyond your participants.
- A biased sample, such as only your friends, would weaken that.
Example
- You plan a study on whether background music harms revision, using a Year 12 group.
- You standardise it: the same room, the same 15-minute test and the same recorded instructions for everyone.
- You operationalise the measure as marks out of 20 on a set vocabulary test, and pilot it on five pupils first.
- The pilot shows the test is too easy, so you make it harder before the full study, protecting validity.
When One Gains and the Other Loses
- Tighter control in a lab usually raises reliability because conditions stay so consistent.
- But a stripped-back lab task can feel artificial, which lowers ecological validity.
- You often trade a little of one to gain the other, so justify where you set the balance.
Common Mistake
- Do not treat reliability and validity as one goal; a study can have one without the other.
- Do not assume more control is always better, because it can cost ecological validity.
- When you suggest an improvement, name which of the two it strengthens.
Method and Design Both Matter
- Choose the method and design with reliability and validity in mind.
- A standardised laboratory experiment with a repeated measures design is easy to repeat, which supports reliability, though its tight control can lower ecological validity.
- Quantitative closed questions are quick to replicate and compare, while qualitative open questions capture richer, more valid detail but are harder to repeat consistently.
Critiquing a Study's Plan
- Check the sampling, the design and the measures in turn.
- Say how each choice affects reliability and validity.
- Suggest one change that would strengthen the weaker of the two.
Self review
- What is the difference between reliability and validity?
- How does standardisation improve reliability?
- Why does a pilot study protect both reliability and validity?
- How can strong lab control lower ecological validity?
- What does an operationalised measure add to a study?