AQA A-level Psychology (7182): Paper 2 – Research Methods – Revision Notes

Research methods in AQA A-level Psychology (7182) is about designing studies, controlling variables, analysing data and judging the quality of findings. It is section 4.2.3 and topic 7 of the AQA AS and A-level Psychology (7181, 7182) specification, Version 1.4 (1 September 2025). It is listed under Paper 2: Psychology in Context (2 hours, 96 marks, 33.3% of the A-level), but research methods and maths skills are assessed on all three papers. These notes are built for fast recall and application to novel scenarios.

For full explanations read the Research methods study guide, then test yourself with the Research methods practice questions.

What the specification lists

1. Methods: lab, field, natural and quasi-experiments; observational techniques; self-report (questionnaires, structured and unstructured interviews); correlations; content analysis; case studies.
2. Scientific processes (4.2.3.1): aims and hypotheses; sampling; pilot studies; experimental designs; observational design; questionnaire and interview design; variables; control; demand characteristics and investigator effects; ethics and the BPS code; peer review; implications for the economy; reliability; validity; features of science; reporting investigations.
3. Data handling and analysis (4.2.3.2): quantitative and qualitative; primary, secondary and meta-analysis; descriptive statistics and calculations; graphs and tables; distributions; correlation coefficients; levels of measurement; coding and thematic analysis.
4. Inferential testing (4.2.3.3): the sign test; probability, significance and critical values; Type I and Type II errors; choosing between the named tests.

Key terms

Term	Definition
Aim	General statement of what is being investigated
Directional hypothesis	Predicts the direction of a difference or relationship
Non-directional hypothesis	Predicts a difference or relationship without direction
Null hypothesis	Predicts no difference or relationship
Operationalisation	Defining a variable in measurable terms
Extraneous variable	Any variable other than the IV that might affect the DV
Confounding variable	An extraneous variable that has affected the DV systematically
Counterbalancing	Varying condition order (e.g. ABBA) to balance order effects
Random allocation	Assigning participants to conditions by chance
Standardisation	Same procedures and instructions for every participant
Demand characteristics	Cues that reveal the aim and change participants’ behaviour
Investigator effects	Researcher’s influence on the result
Pilot study	Small trial run to test the procedure
Event sampling	Recording each occurrence of a target behaviour
Time sampling	Recording behaviour at fixed intervals
Meta-analysis	Statistical combination of many studies’ findings
Peer review	Expert scrutiny before publication
Falsifiability	A theory can in principle be proved wrong
Paradigm	A set of shared assumptions within a science

Designs compared

Design	Strength	Weakness	Fix
Repeated measures	No participant variables between conditions; fewer participants	Order effects (practice, fatigue); may guess the aim	Counterbalancing
Independent groups	No order effects	Participant variables; more participants	Random allocation
Matched pairs	Controls key participant variables; no order effects	Matching is time-consuming and never perfect	Pilot matching variable

Sampling compared

Technique	Representative?	Practical?
Random	Potentially; free of researcher bias	Needs full population list
Systematic	Fairly, if the list has no pattern	Simple once list exists
Stratified	Most representative of sub-groups	Time-consuming
Opportunity	Usually not	Quick and easy
Volunteer	Volunteer bias	Easy; reaches many

Reliability and validity at a glance

Concept	Assess by	Improve by
Test–retest reliability	Same test, same people, two occasions; correlate	Clearer, unambiguous questions
Inter-observer reliability	Two observers; correlate tallies (+0.80 or above)	Operationalised categories; training
Face validity	Does it look right to an expert?	Revise items
Concurrent validity	Compare with an established measure	Use a validated measure
Ecological validity	Do findings apply to real settings?	More natural tasks and settings
Temporal validity	Do findings still hold today?	Replicate in a new era

Descriptive statistics

Measure	Use when	Weakness
Mean	Interval data, no extreme scores	Distorted by outliers
Median	Ordinal data or outliers present	Ignores exact values
Mode	Nominal data	May be several or none
Range	Quick check of spread	Distorted by extremes
Standard deviation	Interval data; precise spread	Harder to calculate

Skew rule. Positive skew: mode < median < mean (tail to the right). Negative skew: mean < median < mode (tail to the left). Normal: all three equal.

Choosing an inferential test

Ask: difference or correlation? Related or unrelated design? Nominal, ordinal or interval data?

	Unrelated	Related	Correlation
Nominal	Chi-squared	Sign test	Chi-squared
Ordinal	Mann-Whitney	Wilcoxon	Spearman’s rho
Interval	Unrelated t-test	Related t-test	Pearson’s r

“Less than” tests: for the sign test, Wilcoxon and Mann-Whitney, the calculated value must be equal to or less than the critical value to be significant. For the others, it must be equal to or greater.

Worked example: the sign test

Ten students rated their mood before and after a mindfulness session. Eight improved (+), one got worse (-) and one did not change (0).

1. Ignore the zero: N = 9.
2. Count the less frequent sign: S = 1.
3. For a directional hypothesis at p ≤ 0.05 with N = 9, the critical value of S is 1.
4. S (1) is equal to or less than the critical value (1), so the result is significant; reject the null hypothesis.

Must-know distinctions

• Lab vs field experiment: setting artificial vs natural; both manipulate the IV.
• Natural vs quasi-experiment: IV varies naturally (event) vs IV is a participant characteristic.
• Covert vs overt: hidden vs known observation.
• Event vs time sampling: every occurrence vs fixed intervals.
• Coding vs thematic analysis: produces numbers vs keeps qualitative themes.
• Type I vs Type II error: false positive (too lenient) vs false negative (too strict).

Frequently asked questions

How many marks is research methods worth? It is Section C of Paper 2, worth 48 of the paper’s 96 marks, and it can be assessed in other papers too.

Which inferential test do I calculate by hand? The sign test. For the others, choose the test from the level of data, the design and whether you are testing a difference or a correlation, then compare the result with a critical value.

What is the difference between a type I and a type II error? A type I error rejects a true null hypothesis (a false positive, more likely with a lenient significance level); a type II error accepts a false null hypothesis (a false negative, more likely with a strict level).

Quick self-test

1. Which design uses counterbalancing?
2. What is the difference between an extraneous and a confounding variable?
3. Name the five sections of a scientific report after the abstract.
4. Which measure of central tendency suits nominal data?
5. In a negative skew, which average is lowest?
6. Which test for an independent-groups design with ordinal data?
7. What is a Type I error?
8. What correlation coefficient shows a strong negative relationship: +0.2, -0.85 or -0.1?

Answers: (1) repeated measures; (2) an extraneous variable might affect the DV; a confounding variable has done so systematically; (3) introduction, method, results, discussion, referencing; (4) mode; (5) mean; (6) Mann-Whitney; (7) rejecting the null hypothesis when it is true (false positive); (8) -0.85.

Where marks are usually lost

• Hypotheses missing the operationalised IV and DV.
• Explaining a sampling technique without applying it to the scenario’s population.
• Saying “results are significant” without comparing calculated and critical values, or using the wrong direction rule.
• Confusing qualitative data with primary data.
• Writing “lacks validity” without saying which type and why.

Official syllabus

AQA AS and A-level Psychology (7181, 7182) specification, Version 1.4 (1 September 2025), section 4.2.3 Research methods — aqa.org.uk. For other topics, see the AQA A-level Psychology hub.