Speca › Topics › GCSE Psychology
Research Methods
AQA GCSE Psychology · Paper 1
Research methods questions turn up in every topic on both papers, so these skills are worth marks everywhere. This page shows how to build each kind of answer one step at a time, from writing hypotheses to working out averages.
Stuck? Worked examples, step by step
How to answer each kind of question in this topic. Cover the model answer, try it yourself, then check each step.
Also on this page: how longer answers are marked, with sample answers at each level · sentence starters
- Writing a null hypothesis
- Random sampling: how, and one weakness
- Spotting an ethical issue and dealing with it
- Naming a design and explaining a weakness in context
- One weakness of a questionnaire, built in four steps
- Planning an answer about correlation and cause
- Mean, median, mode and range, step by step
Write · 2 marks
Writing a null hypothesis
Grace wants to find out whether pupils who drink a glass of water before a test score higher than pupils who do not. Pupils sit a 25-mark quiz. Write a null hypothesis for this study. [2 marks]
First find the two variables. The independent variable (IV) is what the researcher changes or compares: water before the test, or no water. The dependent variable (DV) is what is measured: the quiz score out of 25. Now you can write a sentence that names both.
A null hypothesis predicts no difference. Use the measurement in your sentence, not a vague word such as 'better'. This is called operationalising the variable.
There will be no difference in the scores out of 25 on the quiz between pupils who drank water before the test and pupils who did not. (1)
The second mark is for saying that any difference you find is down to chance. That is what makes it a null hypothesis.
Any difference found will be due to chance. (1)
If the question asked for the alternative hypothesis instead, you would predict a difference and say which way, still using the measurement. For example:
Pupils who drink water before the test will score higher out of 25 on the quiz than pupils who do not.
Why it scores: it names both variables, uses the measurement (score out of 25) and states no difference and chance. 'Water will have no effect on the test' is too vague for full marks.
Apply · 3 marks
Random sampling: how, and one weakness
Aisha wants a random sample of 15 pupils from the 150 pupils in Year 9. She has a list of all 150 names. Explain how she could choose a random sample and give one weakness of random sampling. [3 marks]
Random means every person in the population has an equal chance of being picked. Say how Aisha would make that happen with her list. A hat is the simplest.
Aisha could write each of the 150 names on a slip of paper, put the slips in a hat and draw out 15 names. (1)
Now say why this makes it random. Always use the words 'equal chance'.
Every pupil in Year 9 has an equal chance of being picked. (1)
Now the weakness, and then make it about Aisha's study. A weakness of random sampling is that it can still give an unrepresentative sample by chance.
A weakness is that chance could still give an unrepresentative sample. For example, Aisha might happen to draw mostly boys, so her sample would not be a small copy of Year 9. (1)
Why it scores: the method is clear and could be carried out, 'equal chance' is stated, and the weakness is applied to Aisha. Do not mix this up with an opportunity sample, which is just whoever is available. 'Anyone could have been picked' does not make it random.
Apply · 4 marks
Spotting an ethical issue and dealing with it
Cara tells Year 10 pupils her study is about how well they listen to instructions. In fact she is testing whether pupils make more mistakes on a puzzle when a stranger watches them. She does not tell the pupils or their parents the real aim. Identify one ethical issue in Cara's study and explain how she could deal with it. [4 marks]
Step 1, spot the issue. Ask: what has Cara kept from the pupils? She has hidden the real aim.
The ethical issue is deception, because Cara has misled the pupils about the real aim of her study. (1)
Step 2, link it to the story. Say what she told them and what the truth is. Then say why it matters.
She told them the study was about listening, but it is about whether a stranger watching causes mistakes, so the pupils cannot give informed consent. (1)
Step 3, say how to deal with it. The usual answer is to avoid deception where possible and, if it has to be used, to debrief afterwards.
Cara should avoid deception where she can. If she cannot, she should debrief the pupils afterwards. (1)
Step 4, add detail so the fix is clear: what happens in the debrief.
In the debrief she should explain the true aim and check that no pupil is upset. She should also tell them they can take their data back. (1)
Why it scores: the issue, the link to the story, the fix and its detail are four separate points. A debrief deals with deception after the event, but it cannot undo any distress that has already happened, so do not say it 'removes' the problem.
Identify and explain · 4 marks
Naming a design and explaining a weakness in context
Luca wants to find out whether pupils read faster on paper or on a screen. He pairs up pupils who scored the same on a reading test. One pupil from each pair reads a passage on paper and the other reads the same passage on a screen. He times each pupil. Name the experimental design Luca used and explain one weakness of it in this study. [4 marks]
Step 1, name the design. Ask: are there different pupils in each condition, or the same pupils? There are different pupils, but they have been paired on a score. Different pupils with no pairing would be independent groups; the same pupils twice would be repeated measures. Paired pupils is matched pairs.
Luca used a matched pairs design. (1)
Step 2, pick a weakness of that design and name it. The usual weakness is that matching is hard and never perfect.
A weakness is that matching is never perfect. (1)
Step 3, put it in context. What did Luca match on, and what did he not match on?
Luca only matched the pupils on one reading test score, so the pupils in each pair may still differ in other ways, such as how quickly they read for fun or how used to screens they are. (1)
Step 4, say what that does to his results.
So a difference in reading times could be caused by the pupils in each pair being different, and not by paper against screen. (1)
Why it scores: the design is correctly named, the weakness is explained in terms of Luca's matching and it ends with what that does to the results. 'It takes a long time' is a true weakness but gets less credit when you do not link it to the study.
Evaluate · 3 marks
One weakness of a questionnaire, built in four steps
A psychologist uses a questionnaire to find out how worried pupils feel about exams. Each pupil ticks one of three boxes: never, sometimes, often. Explain one weakness of this questionnaire. [3 marks]
Use four parts: Point (the weakness), Evidence (what in this questionnaire causes it), Explain (why that is a problem), Link back (what it does to the findings). Start with the point. Look carefully at how this questionnaire asks its questions.
One weakness is that closed questions limit what people can say.
Now the evidence from the story. Pupils can only tick one of three fixed boxes.
Pupils can only tick 'never', 'sometimes' or 'often', so they have no way to explain what kind of worry they feel or how strong it is. (1)
Explain why that is a problem. Think about two pupils who tick the same box.
Two pupils who tick 'often' might feel very different amounts of worry, but the questionnaire cannot show this. (1)
Link back to the findings. Questionnaires are often judged on how accurate they are, which is validity.
So the data may not be an accurate picture of how worried pupils really feel, which reduces its validity. (1)
Why it scores: the weakness is specific to a closed questionnaire, backed by a detail from the story and followed through to the findings. Also acceptable, if you explain it in the same way: people may not answer truthfully.
Plan a 6-mark answer
Planning an answer about correlation and cause
Freya finds a negative correlation between the number of lessons pupils miss and their end-of-year exam marks. A friend says, 'So missing lessons causes lower marks.' Discuss whether Freya's friend is right. [6 marks]
'Discuss' means weigh it up, so give more than one side. Plan three short paragraphs: Paragraph 1, say what the correlation does show. Paragraph 2, say what it does not show and why. Paragraph 3, say what a correlation is useful for, and give your judgement.
Paragraph 1 is a short warm-up. For a negative correlation, as one variable rises, the other falls. On a scatter diagram the dots slope downwards. Write one sentence about Freya's data using that idea.
Paragraph 2 is where most marks are won, so write it fully. Say that a correlation shows a link, not cause, and give a third variable that could explain both. Think of something that could make a pupil miss lessons and also do worse in exams.
A correlation only shows that two variables are linked. It does not show that one causes the other, because a third variable could explain both. For example, a pupil who is ill for a long time, or who is unhappy at school, might miss more lessons and also do worse in exams. Freya measured lessons missed and exam marks but did not change either, so she cannot say that missing lessons caused the lower marks. (2)
Paragraph 3 is the judgement. A correlation is a good starting point for research and is useful when the variables cannot be changed, as with missed lessons. So a short ending might be: 'Freya's friend is wrong to say it proves cause, but the link is worth looking into.'
Why it scores: it explains what the correlation shows, explains why it cannot show cause with a specific third variable, and reaches a judgement about the friend. The usual lost marks come from saying 'correlation does not mean causation' with no reason.
Calculate · 6 marks
Mean, median, mode and range, step by step
Isaac gave eight pupils a memory test. Each pupil could score up to 20. Their scores were: 11, 14, 9, 14, 16, 12, 14, 10. Calculate the mean, the median, the mode and the range. [6 marks]
Do the easy bit first: put the scores in order, from smallest to largest. That makes the median, mode and range easy to see. Write the list out.
In order: 9, 10, 11, 12, 14, 14, 14, 16
The mean: add up all the scores, then divide by how many scores there are. Show the adding on one line and the dividing on the next. There are 8 pupils.
Mean: 11 + 14 + 9 + 14 + 16 + 12 + 14 + 10 = 100 (1)
100 ÷ 8 = 12.5 (1)
The median: the middle score when they are in order. There are 8 scores, an even number, so there are two middle scores. The 4th and 5th are 12 and 14. Add them and halve.
Median: (12 + 14) ÷ 2 = 26 ÷ 2 = 13 (2)
The mode: the most common score. Count in the ordered list. 14 appears three times, more than any other score.
Mode = 14 (1)
The range: biggest score minus smallest score.
Range = 16 - 9 = 7 (1)
Why it scores: the scores are put in order first, each step is on its own line, and the median uses the two middle scores because there is an even number of scores. The most common slip is taking the median from the list before ordering it. Check your answer is sensible: the mean and median should sit between the smallest and biggest scores.
Longer answers: structure and levels
Longer questions, worth 4, 6 or 9 marks, are usually marked in levels: the whole answer is read, the level that fits best is chosen, then a mark inside it. These descriptions are Speca's own guide, the same as the one printed in the workbook; the real exam uses the exam board's mark schemes.
How to structure them
| 4 marks | Two points, each explained, or one point described and then explained in detail. If the question has a story in it, use the person's name in every point. |
| 6 marks | Three short paragraphs. To evaluate, use point, evidence, explain, link back in each one, and cover a strength and a weakness. To apply, name the idea, link it to the person in the story, then explain it. Finish with one sentence of judgement if the question asks you to evaluate. |
| 9 marks | Describe first: the theory or study, accurately, in a paragraph. If the question only says evaluate, keep this short. Then two or three evaluation points, each fully explained and using a study as evidence. Cover strengths and weaknesses. End with a judgement that answers the question. Do not spend all your time describing: the higher levels need evaluation as well. |
| Timing | The papers give 1 hour 45 minutes for 100 marks, so about a minute a mark: roughly 9 or 10 minutes for a 9-mark answer, including a minute to plan. |
The levels for a 4-mark answer
| Level | What it looks like |
|---|---|
| 3 to 4 marks | Clear, accurate and detailed. Each point is explained, and where there is a scenario the answer uses it. Psychological terms are used correctly. |
| 1 to 2 marks | Some relevant knowledge, but brief, muddled, or not applied to the question. |
| 0 marks | Nothing relevant. |
The levels for a 6-mark answer
| Level | What it looks like |
|---|---|
| 5 to 6 marks | Several relevant points, each explained clearly and accurately. If the question asks for evaluation, it says why each point matters. If there is a scenario, every point is applied to it. Well organised, with accurate terms. |
| 3 to 4 marks | Relevant points with some explanation, but some are thin, one-sided or only partly applied. |
| 1 to 2 marks | Points are listed or described but not explained, or the answer is mostly off the question. |
| 0 marks | Nothing relevant. |
The levels for a 9-mark answer
| Level | What it looks like |
|---|---|
| 7 to 9 marks | Accurate, detailed description and well-developed evaluation, using a study as evidence. Strengths and weaknesses are both covered, and the answer reaches a judgement. Clearly organised, with accurate terms. |
| 4 to 6 marks | Mostly accurate description with some evaluation, which may be one-sided or not fully explained. A study may be mentioned but not used as evidence. |
| 1 to 3 marks | Limited description, with little or no evaluation, or with errors. |
| 0 marks | Nothing relevant. |
One question, three answers
Mr Lopez wants to find out whether copying out a list of words by hand helps pupils remember them better than reading the list silently. He draws the names of 24 pupils from a hat, using the full Year 9 register. He then randomly allocates 12 pupils to each group. Before the study he sends a letter to parents and tells pupils they can stop at any time. Everyone gets the same 20 words, three minutes to learn them and the same written instructions. The copying group does the study in the morning and the reading group does it in the afternoon. In the test, pupils write down as many words as they can remember. The copying group's mean score is 13 out of 20 and the reading group's mean score is 10 out of 20. Evaluate the method Mr Lopez used. Give strengths and weaknesses, and reach a judgement about how far his result can be trusted. [9 marks]
Read each answer, then the note on why it sits in that level. Try writing your own first, then compare.
Level 1: about 2 marks out of 9
Mr Lopez picked pupils from a hat, which is an opportunity sample because anyone could have been picked. That is good because it is fair. He used the same words and the same instructions so it is a good study. The copying group got more words right so copying works. The study is reliable because it was a good test and measured what it should. It could be better if he had more pupils. It is a good study.
Why this level: This answer has two mistakes: names drawn from a hat give a random sample, where everyone has an equal chance, not an opportunity sample, and 'measured what it should' describes validity, not reliability. It also says copying 'works' when a difference in means is not proof. The strengths and weakness are named in a few words, with no explanation.
Level 2: about 5 marks out of 9
Mr Lopez used a random sample because he drew names from a hat, so every pupil in Year 9 had an equal chance of being picked. This is a strength because there is no researcher bias in who is picked. He also allocated the pupils to the two groups randomly, which helps to make the groups similar. He gave everyone the same words, the same time to learn them and the same instructions, which makes the study easier to repeat. He also asked for the parents' permission and told pupils they could stop at any time, which deals with informed consent and the right to withdraw.
A weakness is that the copying group did the study in the morning and the reading group did it in the afternoon. Pupils might be more tired in the afternoon. There were only 12 pupils in each group, so the sample is quite small.
The copying group did better, with a mean of 13 compared with 10, so copying words might help memory.
Why this level: Most of the evaluation is accurate and applied to Mr Lopez's study, and there are several strengths. But it is one-sided: the weaknesses are brief and not explained, and the answer never says how the time of day could change the result. It ends with a cautious conclusion but does not judge how far the result can be trusted.
Level 3: about 8 marks out of 9
Mr Lopez's study has several strengths. He used a random sample, drawing names from the full Year 9 register, so every pupil had an equal chance of being picked. This avoids researcher bias in who takes part. He used independent groups, with different pupils in each group. This means there are no order effects, such as practice or tiredness, which would be a problem if the same pupils did both conditions. Differences between the people in the groups could still affect the results, but allocating the pupils to groups randomly helps make the groups similar. He also used standardised procedures: the same words, the same learning time and the same written instructions. This makes a difference in scores less likely to be caused by the procedure, and the study could be repeated, which makes it more reliable. Finally, he dealt with informed consent by writing to parents, and with the right to withdraw by telling pupils they could stop at any time.
However, there are weaknesses. The biggest is that the copying group did the study in the morning and the reading group in the afternoon. Time of day is an extraneous variable that was not controlled. Pupils may be more alert in the morning, so the higher mean of 13 compared with 10 might be caused by the time of day and not by copying. Random sampling can also still give an unrepresentative sample by chance, and with only 12 pupils in each group, all from one year group, the result may not apply to other pupils. The task, learning a list of words for three minutes, is also not like real revision, so the results may lack ecological validity.
Overall, the method is good in how it chose and allocated pupils and how it treated them, but the time of day difference means the result cannot be fully trusted. I would trust it more if both groups did the study at the same time of day, and with more pupils.
Why this level: The strengths and weaknesses are accurate and each one is applied to Mr Lopez's study and explained, such as why random allocation matters for independent groups and how the time of day could explain the difference in means. The answer weighs the points up and ends with a clear judgement about how far the result can be trusted, with a suggested improvement.
How to move up a level
- Explain why each point matters, not just what it is. Level 2 says he 'allocated the pupils randomly'. Level 3 adds that this matters because independent groups can differ, and random allocation helps make them similar.
- Develop the biggest weakness. Level 2 says 'pupils might be more tired in the afternoon'. Level 3 names time of day as an uncontrolled extraneous variable and says how it could cause the higher mean of 13, so copying may not be the reason.
- Finish with a judgement and an improvement. Level 3 says how far the result can be trusted and what Mr Lopez should change, such as testing both groups at the same time of day.
Sentence starters
Stuck on how to begin? Pick a starter and finish the sentence in your own words. The first set is for this topic; the second works for any topic.
For this topic
- Describe A random sample is chosen by giving every member of the population an equal chance, for example by...
- Describe Counterbalancing deals with order effects by...
- Explain A correlation does not show cause and effect because a third variable could...
- Explain A study can be reliable without being valid. For example, a bathroom scale that is always 2 kg out is reliable because...
- Apply Dalia writes pupils' real names next to their scores. The ethical issue is confidentiality, and she could deal with it by...
- Evaluate One strength of a field experiment is that behaviour is more natural, which means...
- Evaluate One weakness of opportunity sampling is that it is likely to be biased, because...
- Judgement Overall, a laboratory experiment is a good choice when a researcher wants high control, but its main limit is...
For any topic
- Describe One feature of ... is ...
- Describe In this study, the participants were asked to ...
- Describe The researchers found that ...
- Explain This happens because ..., which means that ...
- Apply In this situation, [name] is showing ..., because ...
- Apply This explains why [name] ..., because ...
- Evaluate One strength of ... is ... This is supported by ..., who found ...
- Evaluate However, one weakness of ... is ... This matters because ...
- Judgement Overall, ... is convincing because ...
- Judgement Overall, ... is limited because ...
- Judgement On balance, the evidence suggests ..., although ...
What this topic covers
- Write an alternative hypothesis and a null hypothesis, with the variables operationalised.
- Identify the independent variable, dependent variable and extraneous variables in a study.
- Describe four sampling methods and give a strength and a weakness of each.
- Describe independent groups, repeated measures and matched pairs designs and evaluate each.
- Tell laboratory, field and natural experiments apart and evaluate each.
- Describe interviews, questionnaires, case studies and observations, including behavioural categories.
- Explain what a correlation is and why it does not show cause and effect.
- Explain standardisation, randomisation, counterbalancing and the control of extraneous variables.
- Explain reliability and validity and how a study can be improved on both.
- Explain six ethical issues and how each can be dealt with.
- Work out a mean, median, mode, range, percentage, fraction and ratio, and use standard form and significant figures.
- Draw and read bar charts, histograms and scatter diagrams, and describe a normal distribution.
Where students lose marks
- Writing a hypothesis that is not measurable ('memory will be better'). Say how each variable is measured, for example 'words recalled out of 20'.
- Calling an opportunity sample 'random' because anyone could have been picked. Random means everyone has an equal chance. Opportunity means whoever is there.
- Saying a correlation proves one thing causes the other. It only shows a link, and a third variable might explain both.
- Mixing up reliability (the same result again) and validity (measuring what it claims to measure).
- Taking the median before putting the scores in order, or forgetting to halve when there are two middle scores.
Key terms
- Hypothesis
- A testable prediction about what a study will find.
- Null hypothesis
- A prediction that there will be no difference or relationship, and that any difference found is due to chance.
- Independent variable (IV)
- The thing the researcher changes or compares.
- Dependent variable (DV)
- The thing the researcher measures.
- Extraneous variable
- Any other factor that could affect the DV if it is not controlled.
- Operationalise
- State exactly how a variable will be measured, for example 'words recalled out of 20'.
- Target population
- The whole group of people the researcher wants to draw conclusions about.
- Sample
- The smaller group of people who are actually studied.
- Order effects
- Changes in performance caused by doing conditions one after the other, such as practice or tiredness.
- Counterbalancing
- Half the participants do the conditions in one order and half in the other, to balance out order effects.
- Behavioural categories
- Clearly defined, observable behaviours that an observer records.
- Inter-observer reliability
- Agreement between two or more observers who record the same behaviour separately.
- Correlation
- A link between two measured variables, which can be positive, negative or absent.
- Reliability
- Consistency: would the same result be found again?
- Validity
- Accuracy: is the study measuring what it claims to measure?
- Ecological validity
- How well the results apply to real life.
- Informed consent
- Participants agree to take part knowing what the study involves.
- Debrief
- Explaining the true aim afterwards and checking that participants are not upset.
- Quantitative data
- Data in the form of numbers.
- Qualitative data
- Data in the form of words or descriptions, not numbers.