Key Idea: A research method is chosen for the question it can answer. Only the experiment answers "what causes this?"; the rest answer "what happens, how often, and what does it mean?"
Change something
- Experiment: set an IV, measure a DV, control the rest
- True: random allocation. Quasi: groups already exist
- The only method that can show cause
Watch or measure
- Observation: record behaviour, change nothing
- Correlational study: relate two variables
- Case study: one case, several methods
Ask
- Survey: the same written questions for many
- Interview: structured, semi-structured, unstructured
- Focus group: a guided group discussion
Who is studied
- Random, stratified: chance decides
- Opportunity, self-selected, snowball: it does not
- Sampling bias limits how far findings apply
The words to know
| Word | What it means |
|---|---|
| Independent variable | What the researcher changes in an experiment |
| Random allocation | Chance decides who goes into which condition |
| Naturalistic observation | Watching behaviour in its everyday setting |
| Inter-rater reliability | How closely two observers' records agree |
| Triangulation | Using more than one source to check the findings |
| Correlation coefficient | A number from -1 to +1: direction and strength of a link |
| Self-reported data | What people tell a researcher about themselves |
| Leading question | Wording that pushes people towards one answer |
| Representative sample | A sample with the same mix as the target population |
Name the method. Say what it can show and what it cannot. Tie both to the study in front of you.
Exam-style question
Your class practical was an experiment. Your class tested whether students recall more of a text after reading it twice than after reading it once and then testing themselves. Compare and contrast the research methodology of the experiment used in your class practical with the research methodology of a focus group.
Model answer plan
See the mark-by-mark plan — for / against / judgement, with marking guidance — in study mode.
Important: Only an experiment with random allocation supports the word causes. For an observation, a survey, a correlation or a case study, write is linked to or suggests.
Tap each card to check yourself.
Which method has an independent variable? Only the experiment: a variable set by the researcher.
Which two methods rely entirely on self-report? The survey and the interview (including focus groups).
Which method collects its data through other methods? The case study.
What does r = -0.8 tell you, and what does it not? A strong negative relationship. Not which variable affects which.
Two observers compare tallies. What are they checking? Inter-rater reliability.
Why is a poster sample not a random sample? People chose to answer it: a self-selected sample, a non-probability technique.
Exam tips
- For any method, know one thing it can show and one thing it cannot.
- Use the method's own words: IV and DV, behavioural categories, open and closed questions, r.
- Say how the sample was found, and who was left out.
- In a comparison, put both methods in the same sentence.