Research Design, Sampling and Ethics
Before you trust a psychology finding, you check how it was made. Learn to spot the variables, choose a design and a sample, and follow the rules that protect the people who take part.
Put it to work, not just to memory
Free and interactive, so you are arguing with the material rather than rereading it.
Start revising freeWhat you'll cover
How to trust a study
A psychology finding is only as good as the way it was produced. Before you believe "X causes Y", you ask three questions: was the research well designed? Was the sample fair? Was it done ethically? This module walks through all three - variables and hypotheses, experimental designs and sampling methods, and the rules that protect the people who take part.
Variables
Every experiment turns on these. The first two are what the study is about; the rest are what can spoil it if left uncontrolled.
Find the variable
Researchers test whether drinking caffeine changes people's reaction time. What is the independent variable?
- Whether or not the participants have caffeine
- The reaction time that is measured
- The temperature of the room
- The age of the participants
Match each variable to an example
- Independent variable
- Dependent variable
- Situational variable
- Participant variable
- The amount of caffeine the researcher gives
- The reaction time that is measured
- Background noise in the testing room
- A participant's age or mood
Null or alternative?
A hypothesis states: "There will be no difference in reaction time between the caffeine and no-caffeine groups; any difference is due to chance." What kind of hypothesis is this?
- A null hypothesis, because it predicts no difference
- An alternative hypothesis, because it predicts a difference
- A directional hypothesis, because it states a direction
- An extraneous variable
Two ways to use your participants
In an experiment you compare conditions, and the design decides WHO does each one. A third design, matched pairs, uses different but closely matched participants to combine the strengths of both.
Beating order effects
In a repeated measures design, participants might improve with practice or tire with fatigue. Select the TWO techniques that control these order effects.
- Counterbalancing - half do condition A first, half do B first
- Randomising the order in which participants do the conditions
- Standardising the instructions given to everyone
- Using a single-blind procedure
Choosing a sample
You cannot test everyone in the target population, so you take a sample. How you pick it decides how representative your results are.
Match each sampling method
- Random sampling
- Stratified sampling
- Volunteer sampling
- Opportunity sampling
- Every person has an equal chance, for example names drawn from a hat
- Subgroups appear in the same proportions as in the population
- People reply to an advert and put themselves forward
- Whoever happens to be available is asked
The rules that protect people
Research uses real people, so it follows ethical guidelines. Learn these five: Informed consent - participants agree knowing what is involved. Deception - do not mislead them; if some is unavoidable, explain fully afterwards. Confidentiality - keep their data private and anonymous. Right to withdraw - they can leave at any time and remove their data. Protection of participants - they should come to no physical or psychological harm.
Run it properly
Put the stages of running an ethical study into a sensible order.
- Decide the aim and write the hypothesis
- Choose the experimental design and the sampling method
- Gain the participants' informed consent
- Run the study with a standardised procedure, allowing withdrawal at any time
- Debrief the participants and keep their data confidential
The ethical call
Three moments in a study. Each time, choose the response that keeps it ethical.
- A researcher wants adults to take part in a memory test. What must they do before it begins?
- Halfway through, a participant says they feel uncomfortable and want to stop. What is the right response?
- The study needed mild deception to work. What must the researcher do afterwards?
Variables and hypotheses
The variable a researcher changes is the _____, and the one they measure is the _____. A _____ hypothesis predicts no difference. In a repeated measures design, order effects are controlled by _____. And the sampling method where people reply to an advert is called _____ sampling.
Design the music and memory study
A psychologist wants to test whether background music affects how many words students can memorise. Design a simple study for this aim, and show how it would be ethical.
- Identify the independent variable and the dependent variable
- Choose an experimental design and a sampling method, with a reason
- Name one extraneous variable you would control, and how you would control it
- Explain two ethical guidelines you would follow, such as informed consent and the right to withdraw