How do psychologists design valid, ethical research?
Apply research methods, variables, sampling and ethics to evaluate psychological studies.
Experimental and non-experimental methods, variables, hypotheses, sampling, validity, reliability and ethical guidelines for evaluating psychological research.
Reviewed by: AI editorial process; not yet individually human-reviewed
Have a quick question? Jump to the Q&A page
What this dot point is asking
Psychology is an empirical science, so it relies on systematic, controlled methods. This dot point is about the toolkit researchers use and how to judge a study's quality.
Research methods
- Experiment: manipulates an independent variable to measure its effect on a dependent variable; the only method that can establish cause and effect. Laboratory experiments have high control; field experiments take place in real settings with more realism but less control.
- Correlational study: measures the relationship between two variables without manipulation. A correlation can be positive, negative or zero, but correlation does not prove causation.
- Observation: watching and recording behaviour, either naturalistic or controlled, overt or covert.
- Case study: an in-depth study of one person or small group (e.g. the amnesiac patient H.M.), rich in detail but hard to generalise.
- Self-report: surveys and interviews gathering people's own accounts, efficient but vulnerable to social desirability bias.
Variables and hypotheses
The independent variable (IV) is manipulated; the dependent variable (DV) is measured. Extraneous variables are unwanted variables that could affect the DV; if they systematically vary with the IV they become confounding variables.
An operational definition states exactly how a variable is measured (e.g. "aggression measured as number of physical contacts in 10 minutes"), making the study replicable.
Experimental design and control
Common designs include independent groups (different participants in each condition), repeated measures (same participants in all conditions), and matched pairs (participants matched on key characteristics). Control techniques include random allocation to conditions, standardised procedures, counterbalancing (to offset order effects in repeated measures), and single- or double-blind procedures.
Sampling
The population is everyone the researcher is interested in; the sample is who actually takes part. Techniques include:
- Random sampling: every member has an equal chance; reduces bias.
- Stratified sampling: the sample reflects subgroups in the population in proportion.
- Systematic sampling: selecting every nth person.
- Convenience (opportunity) sampling: whoever is available; quick but biased.
- Self-selected (volunteer) sampling: people respond to an advert; may attract atypical participants.
A representative sample supports generalisation to the population.
Validity and reliability
Validity is whether the study measures what it claims. Internal validity is whether the IV really caused the change in the DV (threatened by confounds). External validity is whether results generalise to other people (population validity) and settings (ecological validity). Reliability is consistency: would the same procedure produce the same results again? Test-retest and inter-rater reliability are two checks.
Ethics
Australian psychological research follows ethical guidelines (the National Statement and the Australian Psychological Society Code of Ethics). Key principles:
- Informed consent: participants agree knowing the study's nature.
- No undue deception, and debriefing afterward if any deception is used.
- Right to withdraw at any time, including their data.
- Protection from harm (physical and psychological).
- Confidentiality and anonymity of data.
Putting it together
When asked to design a study, state the method, IV, DV, hypothesis, sample and controls, then identify the ethical safeguards. When asked to evaluate a study, work through its validity, reliability, sampling and ethics, giving a balanced judgement rather than a one-sided list.
Exam-style practice questions
Practice questions written in the style of TASC exam questions on this dot point, with worked answer explainers. The year tag is the paper they imitate, not the source.
TCE 20228 marksA researcher tests whether background music affects exam performance. Half a class sits a test in silence and half with music. Identify the IV, DV and one extraneous variable, state a suitable experimental hypothesis, and name and justify the experimental design and one control used.Show worked answer →
This is a research-design item marked on Criterion 3. Work through each element.
- IV and DV
- The IV is the presence or absence of background music (manipulated, two conditions). The DV is exam performance, operationalised as the test score out of the available marks.
- Extraneous variable
- Individual differences in ability between the two groups, or room temperature and time of day, could affect the DV; if one group happened to be more able it would confound the result.
- Hypothesis
- A non-directional experimental hypothesis: there will be a difference in test scores between students who sit the test with background music and those who sit it in silence. (A directional version would predict which group scores higher.)
- Design and control
- This is an independent-groups design because different participants are in each condition. Random allocation to the two conditions controls participant differences by spreading ability evenly, supporting a cause-and-effect claim.
Markers reward correct identification of variables, a testable hypothesis, and a control linked to a named threat.
TCE 202410 marksDiscuss the importance of ethical guidelines in psychological research, referring to at least two principles and one classic study where ethics were questioned. Evaluate whether the scientific value of such research can justify ethical costs.Show worked answer →
This is an extended-response item marked on Criteria 3 and 7. Define principles, apply a study, then evaluate.
- Ethical principles
- Informed consent means participants agree knowing the study's nature and their right to withdraw. Protection from harm means participants should not face physical or psychological harm beyond everyday life, with debriefing to reverse any deception. Australian research follows the National Statement and the APS Code of Ethics.
- Applied study
- Milgram's (1963) obedience study used deception (participants believed they delivered real shocks) and caused visible distress, breaching protection from harm and full informed consent. Participants were debriefed and most later reported being glad to have taken part.
- Evaluation
- The scientific value was high: the findings transformed understanding of obedience to authority. Against this, the distress caused and the compromised consent are serious costs, and the study could not be run unchanged today under current guidelines. A balanced judgement is that ethical safeguards (consent, debriefing, harm minimisation) must constrain research, and that value alone cannot license avoidable harm, though debriefing and the lasting insight partly offset the costs here.
Markers reward defined principles, an apt study, and a genuinely two-sided evaluation rather than a one-line verdict.