Research Methods and Scientific Inquiry
Core concepts
You must understand the research process: formulating a hypothesis, operationalising variables, choosing a design, sampling, controlling extraneous variables, analysing data and drawing conclusions. Know experimental designs (independent groups, repeated measures, matched pairs) with strengths and weaknesses, and non-experimental methods (correlational, observational, case study, survey, quasi-experiment).
Know variables: independent, dependent, extraneous, confounding. Understand reliability (consistency) versus validity (internal, external, construct, ecological). Know demand characteristics, investigator effects, participant variables, order effects and counterbalancing, random allocation versus random sampling.
Ethics are examinable: informed consent, deception, protection from harm, confidentiality, right to withdraw, and debriefing. Know descriptive statistics (mean, median, mode, range, standard deviation) and when each is appropriate, plus basic inferential logic: probability, significance level, Type I and Type II errors. Correlation does not establish causation.
Common traps
- Confusing random sampling (who is chosen) with random allocation (who goes to which condition).
- Claiming a significant result proves the hypothesis; it indicates the result is unlikely under the null.
- Confusing reliability with validity; a scale can be consistently wrong.
- Forgetting that order effects are specific to repeated-measures designs.
Rule of thumb
For any study, ask: what was manipulated, what was measured, what was held constant, and who were the participants? Then evaluate validity, reliability and ethics in that order.
Worked example
A researcher tests whether sleep improves test scores. Independent variable: hours of sleep (manipulated). Dependent variable: test score (measured). Confound to control: prior knowledge, controlled by random allocation and standardised testing conditions.