Methods of Enquiry in Psychology
Methods are a high-yield area: the exam asks you to match a research design to its purpose, read a correlation, and apply ethical safeguards.
Observation
Naturalistic observation studies behaviour where it normally occurs; controlled observation sets conditions. The observer may be participant or non-participant. Inter-rater reliability and observer bias are key concerns.
Survey
Questionnaires and interviews collect self-reports from a sample. Random or stratified sampling improves representativeness. Wording, social desirability and response sets distort data.
Experiment
The independent variable is manipulated, the dependent variable measured, extraneous variables controlled and participants randomly assigned. Only this design supports causal claims.
Case study
An in-depth study of one person or small group, often clinical. Rich detail, but weak generalisability.
Correlation
The coefficient r ranges from -1 to +1. The sign gives direction; the magnitude gives strength. Correlation does not establish causation because a third variable may drive both.
Ethics
Informed consent, no harm, confidentiality, right to withdraw, debriefing after deception and avoiding coercion.
Worked example
A study reports r = -0.62 between hours of social media use and sleep quality. Reasoning: the negative sign means more use goes with poorer sleep; 0.62 is a moderate-to-strong association. It still does not prove causation because stress could cause both. Only an experiment manipulating social media use could test cause and effect.
Traps
- A negative correlation is inverse, not necessarily weak.
- A survey comparing existing groups is not an experiment; nothing was manipulated.
- Random sampling (who is studied) differs from random assignment (who is in each condition).