How do psychologists design valid, reliable and ethical investigations of behaviour?
Apply research methods including experimental design, variables, sampling, validity, reliability and the interpretation of descriptive and inferential statistics.
A focused answer to the WACE Year 12 Psychology research-methods dot point taught across both units. Covers variables, experimental design, sampling, extraneous and confounding variables, validity, reliability, and descriptive versus inferential statistics with worked examples.
Reviewed by: AI editorial process; not yet individually human-reviewed
Have a quick question? Jump to the Q&A page
Jump to a section
What this dot point is asking
This dot point asks you to design, evaluate and interpret psychological investigations. It is examined throughout the course, so master the vocabulary precisely.
Variables
The independent variable (IV) is the variable the researcher manipulates. The dependent variable (DV) is what is measured to see the effect. Operationalisation means defining variables in measurable terms (for example, defining "stress" as score on a validated anxiety questionnaire).
Experimental designs
- Independent groups: different participants in each condition. Quick and avoids order effects, but individual differences between groups can confound results.
- Repeated measures: the same participants in every condition. Controls individual differences, but introduces order effects (practice or fatigue), which counterbalancing can reduce.
- Matched participants: participants are paired on a relevant characteristic, then one of each pair is allocated to each condition, combining advantages of the other two designs.
A true experiment also uses a control group and random allocation of participants to conditions.
Sampling
The population is everyone the researcher wants to generalise to; the sample is the subset studied.
- Random sampling: every member of the population has an equal chance of selection, reducing bias.
- Stratified sampling: the population is divided into strata (for example, year groups) and sampled in proportion, improving representativeness.
- Convenience sampling: using readily available participants; fast but often biased and less representative.
Validity and reliability
Validity is whether a study measures what it claims to.
- Internal validity: the degree to which the IV (not confounds) caused the change in the DV.
- External validity: how well results generalise to other people and settings (ecological validity).
Reliability is consistency. Internal consistency checks whether items in a test measure the same thing; test-retest reliability checks stable results over time; inter-rater reliability checks agreement between observers. A measure can be reliable without being valid (consistent but measuring the wrong thing), but a valid measure must be reliable.
Descriptive statistics
Descriptive statistics summarise data.
- Measures of central tendency: mean, median and mode. The mean is sensitive to outliers; the median is more robust for skewed data.
- Measures of spread: range and standard deviation. The standard deviation shows how far scores typically deviate from the mean; a larger SD means more variability.
Inferential statistics
Inferential statistics test whether results are likely due to the IV or to chance. Researchers set a significance level, conventionally p < 0.05, meaning a result this extreme would occur by chance less than 5 percent of the time if the null hypothesis were true. If the obtained probability is below 0.05, the result is statistically significant and the null hypothesis is rejected.
In the exam, always state the IV, DV and how each is operationalised, identify likely confounds, and interpret p-values against the 0.05 threshold.
Exam-style practice questions
Practice questions written in the style of SCSA exam questions on this dot point, with worked answer explainers. The year tag is the paper they imitate, not the source.
WACE 20216 marksA researcher tests whether listening to music while studying affects recall. Participants study a word list either in silence or with music playing, then complete a recall test. Identify the independent and dependent variables, name two extraneous variables that should be controlled, and state one way to control each.Show worked answer →
A 6 mark research-method response needs the variables operationalised plus two controlled extraneous variables.
Variables. The independent variable is the study condition (music versus silence), which the researcher manipulates. The dependent variable is recall, operationalised as the number of words correctly recalled.
Extraneous variables and controls. Time of day or fatigue could affect recall; control it by testing all participants at the same time of day. Prior familiarity with the words could affect recall; control it by using a list of unrelated nonsense words or random words none of the participants has seen.
One mark each for the IV, the DV, two named extraneous variables, and a matched control for each. Markers reward operationalisation (saying how recall is measured) rather than vague variable names.
WACE 20235 marksA study reports that a new revision technique produced a mean test score of 72 percent compared with 61 percent for the standard technique, with the difference returning a value of . Explain what the p-value means and whether the result is statistically significant, and explain one limitation of relying on statistical significance alone.Show worked answer →
A 5 mark answer needs the meaning of the p-value, the significance judgement, and one limitation.
- Meaning
- The p-value is the probability of obtaining a difference this large by chance if the null hypothesis (no real difference) were true. Here means there is a 3 percent probability the result is due to chance.
- Significance
- Because is below the conventional threshold, the result is statistically significant, so the researcher rejects the null hypothesis and concludes the revision technique improved scores (assuming confounds were controlled).
- Limitation
- Statistical significance does not show the effect is large or meaningful in practice. An 11 percentage-point gain may or may not matter educationally, and significance can be reached with a large sample even for a trivial effect. Markers reward distinguishing significance from effect size.