Why methods matter
Psychology is a science because its claims are tested with evidence. Everyday impressions ("my friend works better at night") are easy to fool. Research methods are the rules that keep us honest.
Variables and definitions
A variable is anything that can change. To measure something fuzzy like "stress", a researcher gives an operational definition: a precise statement of how it will be measured, such as a score on a 20-item questionnaire. Without that, two studies could not be compared.
Descriptive and correlational research
Descriptive methods (surveys, observation, case studies) tell us what is happening. A correlational study measures whether two variables move together. The correlation coefficient r runs from −1 to +1:
- Near +1: as one rises, the other rises.
- Near −1: as one rises, the other falls.
- Near 0: no linear relationship.
The size shows strength; the sign shows direction. A correlation of −0.8 is strong and negative.
The key warning: correlation does not prove causation. Ice cream sales and drownings both climb in summer, but ice cream does not cause drowning. A third variable (hot weather) drives both.
Experiments
Only an experiment can support a causal claim. The researcher:
- Manipulates the independent variable (IV), the factor being tested.
- Measures the dependent variable (DV), the outcome.
- Uses random assignment to place participants in groups by chance, so the groups start out equal on average.
- Compares an experimental group with a control group that does not get the treatment.
Other tools reduce bias. A placebo is an inactive treatment. In a double-blind study neither the participants nor the staff collecting data know who is in which group, so expectations cannot steer the results.
Worked example
A researcher thinks caffeine improves memory. She recruits 80 students and randomly assigns 40 to drink coffee and 40 to drink decaf, without telling them or the person scoring the tests which is which. Everyone then studies a word list and is tested.
- IV: caffeinated or decaf drink.
- DV: number of words recalled.
- Control group: the decaf drinkers (a placebo).
- Random assignment guards against, say, all the morning people landing in one group.
- Double-blind design prevents expectation effects.
If the coffee group recalls significantly more words, she can conclude caffeine caused the difference, within the limits of her sample.
Sampling and ethics
A sample is a subset of a population. A random sample, where every member has an equal chance of selection, is the best way to make results generalise.
Ethics rules require informed consent, protection from harm, confidentiality and debriefing after any deception. University review boards approve studies before they begin.