Component 2: Principles of research — Eduqas A-Level Psychology
Test yourself on Component 2: Principles of research with EDUQAS A-Level practice questions.
Start free
7 days Premium · Then the free plan continues · No card, no charge
Your focus
- Learners should appreciate the limitations of scientific research and when dealing with the complexities of humans as test material, there are several issues which need to be considered.
- In order to contextualise some aspects of research methods learners are required to consider the methodologies used by both social and developmental psychology
Component 2: Principles of research exam tips
Quick Revision Summary (Key Takeaway)
Component 2: Principles of research in Eduqas A-Level Psychology assesses scientific methodology, experimental design, descriptive and inferential statistics, and ethical considerations. Mastery requires understanding how to design novel research, select and interpret statistical tests, and critically evaluate both internal and external validity in psychological investigations.
Topic Overview
Eduqas Component 2 (Section B: Principles of Research) is designed to evaluate your operational grasp of the scientific method within psychological inquiry. You will be tested on the mechanics of experimental and non-experimental methodologies, sampling techniques, operationalisation of variables, and the elimination of extraneous variables such as demand characteristics and investigator effects.
Furthermore, this component heavily prioritises data handling, requiring both quantitative literacy (descriptive measures, levels of measurement, inferential statistics) and qualitative analysis (thematic and content analysis). Mastering these research principles is essential because they form the methodological backbone for evaluating psychological theories and classical research studies throughout the entire A-Level qualification.
Key Concepts
- →Levels of Measurement: Distinguishing between nominal (categorical), ordinal (ordered ranks with unequal intervals), and interval/ratio (standardised units with equal intervals).
- →Inferential Statistics: Selecting tests (e.g. Mann-Whitney, Wilcoxon, Spearmans Rho, Chi-Square, Unrelated/Related t-tests, Pearsons r) using the criteria of test purpose, design type, and data level.
- →Validity and Reliability: Operationalising mechanisms to assess and enhance both internal validity (standardisation, double-blind controls) and external validity (ecological, temporal, population), alongside reliability assessments (test-retest, split-half, inter-rater).
- →Scientific Processes: Formulating null and alternative (directional or non-directional) hypotheses, operationalising independent and dependent variables, and applying peer review principles.
Examiner Tips
- 💡Contextualise all answers! When asked how to improve validity or select a sample, generic answers gain a maximum of Level 1 marks. Quote the specific variables, apparatus, and participants from the stem.
- 💡Memorise the exact decision rules for critical values. State clearly: 'For [Test Name], the observed value must be [less than or equal to / greater than or equal to] the critical value', before comparing the numbers.
- 💡When explaining how to deal with ethical issues, always use a two-pronged approach: state the mitigation method (e.g. debrief, presumptive consent) and explain specifically how it resolves the dilemma in that exact study.
Common Mistakes
- Believing that a 'significance level of p <= 0.05' means there is a 95% probability that the hypothesis is true. In reality, it means there is a 5% or lower probability that the observed results occurred due to chance alone if the null hypothesis were true.
- Assuming that qualitative data cannot be converted into quantitative data. Content analysis directly operationalises qualitative materials into quantitative frequency tallies through clear coding categories.
- Thinking all inferential tests require the observed value to be greater than the critical value. Tests containing the letter 'R' in their name (e.g. Spearmans Rho, Pearsons r, Chi-Square) require observed >= critical, whereas tests without an 'R' (Mann-Whitney U, Wilcoxon T, Sign test) require observed <= critical.
Revision Plan
- 1Step 1: Master the foundational vocabulary (independent/dependent variables, operationalisation, reliability vs validity types, and directional vs non-directional hypotheses).
- 2Step 2: Create a decision matrix for inferential tests. Practice justifying tests across 10 novel scenarios using the three-part rule (aim, design, measurement level).
- 3Step 3: Drill data analysis skills: calculate percentage changes, identify modal/median/mean values, sketch distributions (normal, positive, and negative skews), and interpret critical value tables.
- 4Step 4: Practice 10-12 mark scenario design questions under timed conditions, ensuring full coverage of sampling, apparatus, operationalised procedures, control of extraneous variables, and ethical safeguards.
Exam Question Types
- 📋Data Interpretation & Calculations (2-4 marks): Calculating percentage differences, mean/median/mode, or reading inferential tables and stating hypothesis decisions.
- 📋Short Application / Methodological Critique (3-6 marks): Identifying extraneous variables in a novel stem and proposing precise methodological modifications to improve reliability or validity.
- 📋Designing Novel Investigations (10-12 marks): Planning a complete psychological study from scratch based on a short scenario prompt, covering hypothesis, design, variables, controls, sampling, and ethics.
Command Word Expectations (EDUQAS)
Justify
Provide clear, evidenced psychological reasons for selecting a specific method, design, or statistical test. Must link the criteria directly to the context provided in the question.
Explain
Clarify how or why a psychological phenomenon, methodological flaw, or ethical issue operates. Requires cause-and-effect reasoning with specific terminology.
Design
Construct an operational, replicable research procedure addressing all specified parameters (e.g. sampling, materials, ethics, controls) in the context of the stem scenario.
How Students Lose Marks (Examiner Pitfalls)
Pitfall: Confusing nominal data with ordinal data when selecting an appropriate inferential statistical test, or failing to justify test selection across all three required criteria.
❌ Weak Answer (Loses Marks):The researcher should use a Chi-Square test because it tests for a difference and uses independent groups.
Example improved answer:The researcher should use the Chi-Square test of association. Firstly, the study is testing for a difference or relationship between two categorical variables. Secondly, the experimental design uses independent groups (unrelated data). Thirdly, the level of measurement is nominal, as participants are placed into mutually exclusive categories (e.g. passed or failed).
Examiner Tip: Always explicitly state three criteria when justifying inferential tests: the hypothesis type (difference vs relationship/correlation), the experimental design (related vs unrelated), and the exact level of measurement (nominal, ordinal, or interval/ratio).
Pitfall: Providing generic definitions of ethical issues without contextualising how to manage or minimise the risk within the novel scenario provided.
❌ Weak Answer (Loses Marks):The psychologist must follow ethics by getting informed consent and giving participants the right to withdraw from the experiment.
Example improved answer:To address informed consent regarding memory performance under stress, the psychologist must provide a comprehensive information sheet detailing the nature of the cognitive task before testing begins. Participants must explicitly sign a consent form acknowledging their voluntary participation, and be informed that they can withdraw their data at any stage up until publication without penalty.
Examiner Tip: Never write generic ethical advice. Directly reference the specific procedures, tasks, or materials described in the exam prompt to demonstrate application.
Step-by-Step Worked Solutions
Question: A psychologist investigated the effect of caffeine on reaction times (in milliseconds). The mean reaction time for the caffeine group was 320 ms (SD = 14.2), and for the decaf group was 365 ms (SD = 42.8). Calculate the percentage difference in mean reaction time between the decaf and caffeine groups relative to the decaf baseline. Show your workings.
- 1.Step 1: Identify baseline and comparison values. Baseline (decaf mean) = 365 ms, comparison (caffeine mean) = 320 ms.
- 2.Step 2: Calculate the absolute difference between the two means: 365 - 320 = 45 ms reduction.
- 3.Step 3: Divide the absolute difference by the original decaf baseline value: 45 / 365 = 0.12328...
- 4.Step 4: Multiply by 100 to convert to a percentage and round appropriately: 0.12328... * 100 = 12.33% (to 2 decimal places).
Final Answer: The caffeine group showed a 12.33% faster mean reaction time compared to the decaf baseline group.
Question: A researcher calculated an observed value of Wilcoxon T = 14 for a directional hypothesis with N = 18 at p <= 0.05. The critical value table states that for N = 18 at p <= 0.05 (one-tailed), the critical value of T is 40. Explain whether the researcher should accept or reject the null hypothesis.
- 1.Step 1: Identify the decision rule for the Wilcoxon signed-ranks test: The calculated/observed value of T must be less than or equal to the critical table value to be statistically significant at the stated significance level.
- 2.Step 2: Compare the observed value to the critical value: Observed T = 14, Critical T = 40. Since 14 is less than or equal to 40 (14 <= 40), the result is statistically significant at p <= 0.05.
- 3.Step 3: Conclude regarding hypothesis status: Because the result is statistically significant, the researcher must reject the null hypothesis and accept the alternative (directional) hypothesis.
Final Answer: The researcher should reject the null hypothesis because the observed value of T (14) is less than the critical value (40), indicating statistical significance at p <= 0.05.
Active Recall Memory Test
What three criteria must you state to justify using a Mann-Whitney U test?
Key Fact: 1. Testing for a difference. 2. Independent groups design (unrelated data). 3. Data is at ordinal level (or interval treated as ordinal).
What is the key difference between concurrent validity and predictive validity?
Key Fact: Concurrent validity validates a measure against an established test given at the same time; predictive validity evaluates how accurately a measure predicts future outcomes or performance.
How do you control for order effects in a repeated measures design?
Key Fact: Use counterbalancing (e.g. an ABBA or split-group design where half complete condition A then B, and half complete B then A).
What direction does the tail point in a positively skewed distribution, and where does the mean sit relative to the median?
Key Fact: The long tail points to the right (positive end); the mean is pulled toward the tail and is higher than the median.
Frequently Asked Questions
How do I know whether data is nominal, ordinal, or interval?
Nominal data consists of headcounts in discrete, mutually exclusive categories (e.g. smoker vs non-smoker). Ordinal data represents ranked or subjective ratings where the intervals between numbers are not mathematically equal (e.g. rating anxiety on a scale of 1-10). Interval data uses standardised, fixed units of measurement with equal intervals between each point on the scale (e.g. time in seconds, temperature in Celsius, height in centimetres).
Do I need to calculate statistical formulas like Wilcoxon or Chi-Square by hand in the exam?
No. In Eduqas Component 2, you will never be asked to compute full inferential test statistics from raw data tables. You are required to identify which test to use, justify your selection, calculate simple descriptive statistics (mean, median, mode, percentages), and compare a provided calculated/observed value with a critical value table to reach a conclusion regarding the hypothesis.
What is the difference between a Type I and a Type II error?
A Type I error (false positive) occurs when a researcher rejects a true null hypothesis, concluding there is a significant effect or difference when one does not exist; this often happens when the significance level is set too leniently (e.g. p <= 0.10). A Type II error (false negative) occurs when a researcher accepts a false null hypothesis, concluding there is no significant effect when one actually exists; this typically occurs when the significance level is set too stringently (e.g. p <= 0.01).
How should I structure a 10-12 mark 'Design an investigation' question?
Organise your response using clear subheadings: 1. Operationalised Experimental Hypothesis (directional or non-directional) and Null Hypothesis; 2. Experimental Design & Variables (IV, DV, operational definitions, and counterbalancing/matching if applicable); 3. Sampling Method (target population, sampling technique, and recruitment steps); 4. Standardised Procedure & Materials (step-by-step instructions so another researcher can replicate it); 5. Control of Extraneous Variables (standardised instructions, single/double blind controls); and 6. Ethical Considerations (brief mitigation of consent, confidentiality, protection from harm).
Why is peer review so important in psychological research?
Peer review involves independent experts assessing research manuscripts for scientific validity, methodological robustness, originality, and ethical compliance before publication. It acts as a quality gatekeeper to prevent fraudulent or flawed findings from entering the scientific record and informs funding allocation decisions by scientific bodies.