Master the OCR A Level Psychology (H567) exams with our comprehensive guide. Learn how to avoid critical marks loss on Paper 1 research designs, accurately operationalise variables, structure top-tier Paper 2 debates, and apply the correct significance rules for statistical tests like the Mann-Whitney U.
閱讀時間 4 分鐘更新於: 2026年6月21日
試卷概覽
卷數
3
總分
300
考試時間
6小時
題型
4
試卷
時間
分數
題數
比重
題型
Research methods
2小時
90
33
30%
選擇題, Short Answer / Hypothesis formulation, Design & Justification Essay, Scenario-Based Short Answer, Calculation and Workings, Graph Plotting, Short / Medium Answer
Psychological themes through core studies
2小時
105
11
35%
Core Study Short Description, Structured Description and Link, Usefulness Essay, Article Link & Evaluation, Evaluating Suggestions Essay
Applied psychology
2小時
105
8
35%
Historical Views & Registry Analysis, Non-Biological Treatment Application, Social Sensitivity Discussion, Key Study and Brain/Police Application, Methodological/Ethical Debate Essay, Scenario Practical Suggestion
評級
A*ABCDEU
計算機規定
A scientific or graphical calculator that meets JCQ regulations may be used (some GCSE Mathematics and Science papers are non-calculator). Graphical calculators must be set to exam mode; you must clear any stored programs, notes or data before the exam, and the calculator must not be able to retrieve stored text or formulae.
用途: Plot a function to read its roots (zeros), points of intersection, and maxima/minima.
使用時機: Checking solutions, sketching, or solving where an analytic method is hard.
步驟
Graph the function(s) and use the built-in zero, intersect and maximum/minimum tools.
考試提示: Allowed under JCQ rules, but you must still show your method — an unsupported calculator answer earns no method marks. Clear all stored programs, notes and data (graphical calculators in exam mode) before the exam.
Numerical equation solver
Graphical calculator / GDC (exam mode)
用途: Solve an equation or find a variable numerically when an algebraic route is long or implicit.
使用時機: Iterative or implicit equations, or to confirm an algebraic solution.
步驟
Use the equation/zero solver, entering the equation and a sensible starting estimate.
考試提示: Allowed under JCQ rules, but you must still show your method — an unsupported calculator answer earns no method marks. Clear all stored programs, notes and data (graphical calculators in exam mode) before the exam.
Numerical integration & differentiation
Graphical calculator / GDC (exam mode)
用途: Evaluate a definite integral \(\int_a^b f(x)\,dx\) or a gradient \(f'(x)\) at a point.
使用時機: Checking calculus answers, or where only a numerical value is needed.
步驟
Use the GDC's numeric integral / derivative function with the limits or the point.
考試提示: Allowed under JCQ rules, but you must still show your method — an unsupported calculator answer earns no method marks. Clear all stored programs, notes and data (graphical calculators in exam mode) before the exam.
Statistics & probability distributions
Graphical calculator / GDC (exam mode)
用途: 1-var/2-var statistics, linear regression, and cumulative binomial / normal / Poisson probabilities without tables.
使用時機: Statistics questions and hypothesis tests.
步驟
Enter data in the statistics editor, or use the distribution menu (binomial cdf, normal cdf, …).
考試提示: Allowed under JCQ rules, but you must still show your method — an unsupported calculator answer earns no method marks. Clear all stored programs, notes and data (graphical calculators in exam mode) before the exam.
常見錯誤
1high涉及分數: 3Research methods and techniques
Failing to fully operationalise the Dependent Variable in Section B, research design (e.g., using 'level of concentration' instead of detailing a quantitative test score like 'spot-the-difference score out of 20').
如何避免: Always state the exact tool, task, or measurement scale used to gather numerical data for the DV.
2medium涉及分數: 4Research methods and techniques
Mislabeling a 1-10 rating scale as a Likert scale, which automatically caps the maximum marks achievable for the research design task.
如何避免: A Likert scale must measure agreement levels (e.g., strongly agree to strongly disagree). If using simple numbers, call it a 'numerical rating scale' instead.
3high涉及分數: 4Practical activities
Failing to make explicit, contextualized links to own practical experiences when designing a study in Paper 1.
如何避免: Explicitly write about your own classroom activities (e.g., 'In my own practical activity on memory, I used...') to prevent your score being capped at 11/15.
4high涉及分數: 2Data recording, analysis and presentation
Omitting the word 'Mean' in the title or the y-axis label when constructing a bar chart in Paper 1.
如何避免: Ensure your y-axis reads 'Mean number of [variable]' and your title explicitly states 'Bar chart showing the mean number of...'
5medium涉及分數: 4Issues in mental health
Proposing biological treatments (e.g., drug therapy, SSRIs) in Paper 3 Q3 when the prompt specifically demanded a non-biological intervention, leading to zero marks.
如何避免: Read the prompt carefully. Non-biological treatments include Cognitive Behavioural Therapy (CBT), systematic desensitisation, or flooding.
6medium涉及分數: 2Data recording, analysis and presentation
Assuming that the calculated value of U in a Mann-Whitney test must be larger than the critical value to be significant.
如何避免: Remember that for a Mann-Whitney U test, the calculated value of U must be equal to or less than the critical value to show significance.
7medium涉及分數: 4Report writing
Double-crediting details in Paper 2 Q11e and Q11f by repeating procedural suggestions in the evaluation question instead of applying distinct psychological criteria.
如何避免: Ensure suggestions are described in part (e), while part (f) focuses strictly on evaluation concepts (ethics, validity, usefulness, reliability).
8high涉及分數: 3Research methods and techniques
Failing to state both variables in a two-tailed alternative hypothesis, or proposing a two-tailed hypothesis when a one-tailed is specified.
如何避免: Ensure both the IV (both conditions) and DV are clearly operationalised, and ensure the direction is specified if a one-tailed hypothesis is requested.