What the exam asks
- Method identification (AAQ Part A & MCQ): Distinguishing between experiments, correlational studies, case studies, naturalistic observations, and meta-analyses from a short study summary.
- Variables and hypotheses (AAQ Part B & MCQ): Identifying the variables of interest, stating the non-experimental hypothesis (a prediction of a relationship, not a causal effect), and writing measurable operational definitions.
- Limits of correlation (MCQ & EBQ): Explaining why correlation does not establish causation by applying the directionality problem or the third-variable problem.
- Data collection and bias (MCQ): Distinguishing surveys (a measurement technique) from research designs, identifying self-report bias, social desirability bias, and wording effects, and classifying measures as qualitative or quantitative.
Core ideas
Comparing non-experimental research methods
Unlike an experiment (which manipulates an independent variable and uses random assignment to establish cause and effect), non-experimental methods observe or measure variables as they naturally occur.
| Research Method | What the Researcher Does | Best Used When... | Key Limitation |
|---|---|---|---|
| Correlational study | Measures two or more naturally occurring variables to see if they co-vary ( from to ). | Predicting one variable from another when manipulation is unethical or impossible. | Cannot establish causation (directionality and third-variable problems). |
| Naturalistic observation | Records behavior in its natural environment without intervention or manipulation. | Studying authentic, everyday behavior (e.g., playground interactions). | No control over variables; observer presence may alter behavior if noticed. |
| Case study | Conducts an in-depth investigation of a single individual or very small group. | Studying rare phenomena (e.g., specific brain lesions or unusual disorders). | Findings cannot be generalized to the broader population; cannot show causation. |
| Meta-analysis | Statistically combines and analyzes the quantitative results of many previous studies on one topic. | Determining the overall effect size or consensus across conflicting literature. | Quality depends on the quality of the included studies (publication bias). |