Your four class practicals are worth 20 marks on Paper 2, and they are the only marks in the whole course that you can fully prepare in advance.
Every question in Section A comes back to them. What method you used, how you used it, one concept applied to it, how it compares with another method, and how you would investigate the same topic differently.
Which means the single highest-value habit in the first year of this course is writing each practical up properly the week you do it. The version of the practical you remember in May will be the version you wrote down now.
What a class practical is
A small, safe, teacher-supervised investigation that you carry out during the course to learn a research method from the inside.
You will do four, covering different methods. The point is not the finding. The finding will be small and your sample will be tiny. The point is that you understand the methodology well enough to describe it precisely and evaluate it honestly months later.
A minimum of around 10 participants is workable for a student-led class practical. More is better where it is practical.
Which method goes with which context
The standard pairings are interview with health and well-being, observation with human development, survey with human relationships, and experiment with learning and cognition.
These are not the only permitted combinations. A school may use a method in a different context, provided you engage with all four research methods. So if your class paired them differently, nothing changes about what you need to record or how Section A will question you.
What each practical is really training
While you are running it, you are also building the material for four different exam questions.
| While you do the practical | You are preparing for |
|---|---|
| Use the method accurately rather than just completing an activity | 1a: detailed knowledge of how the method was applied |
| Notice methodological problems while they are happening | 1b and Section B: concept application with real consequences |
| Record enough detail to reconstruct the procedure months later | 1a and 1d: accurate procedure under exam conditions |
| Ask what the evidence can and cannot show | 1c and Section B: comparison, evaluation and calibrated conclusions |
| Reflect immediately after data collection | A permanent record that makes revision much easier |
The ethics rules specific to class practicals
These are more specific than general ethical principles, and two of them are not judgement calls.
Conformity and obedience studies are not permitted as class practicals under any circumstances. Animals must not be used.
Sensitive topics are also out. Do not ask participants about diagnoses, self-harm, bullying, trauma or other sensitive personal experiences. If you want to run an interview practical, choose something low-risk like everyday study pressure and coping strategies.
On consent, the rules are precise.
| Situation | What is required |
|---|---|
| Participants aged 16 and over | Their own explicit written informed consent. Parental consent is not required |
| Participants under 16 | Written consent from a parent or guardian, as well as the participant's own agreement, with parents properly informed about what taking part involves |
| Participants under 12 | Permitted, provided ethics are handled carefully, with particular attention to parental informed consent |
| Any practical with children in a school | Written consent of the teachers concerned is also required |
| Any class practical at all | Explicit consent through a consent form. Implied consent is not acceptable |
The naturalistic observation exception. In a naturalistic observation, participants cannot give informed consent, because asking would alert them to being observed and change the behaviour you are trying to record. Debriefing is usually impractical too. The accepted position is that observation is acceptable in places where people would expect to be seen by strangers, such as a shopping centre, public transport or the street. Your teacher decides what is appropriate.
Debriefing is not uniform across designs. It is an absolute requirement where information was withheld or partial deception was used, and in that case participants keep the right to withdraw their data at the debrief. After other designs it is often good practice and sometimes expected, but the strength of the requirement follows from the design.
The ethics tension worth being able to argue
This is where ethics stops being a moral add-on and becomes methodology, which is how it is examined.
Full disclosure protects autonomy but invites demand characteristics. Mild deception protects internal validity but costs informed consent.
A top-band answer does not simply condemn deception. It weighs the trade-off and says what the study gains and loses. Something like: not telling participants the true aim protected the study from demand characteristics, since participants who knew what was being measured would probably have behaved differently. The cost is that consent was not fully informed, so it depended on a thorough debrief and a genuine opportunity to withdraw data. Given the task carried little risk of harm, the deception is defensible, but the study's credibility rests on the quality of that debrief.
Notice what that does. It names the gain, names the cost, and reaches a judgement.
What to record, and when
Do this while it is still fresh. Not in May.
Every practical needs the same core record: the research question and aim in one sentence, the method and its specific subtype, the target population and actual sample with size, the sampling technique, the variables or constructs, exactly what counted as the behaviour or score, the procedure in six to eight precise steps, what was actually recorded, and the ethics steps you took.
Then there are method-specific fields that are the ones people forget first.
| If you ran an... | Also record |
|---|---|
| Experiment | Setting, true or quasi and the design used, IV and DV operationalised with the actual values and scoring method, allocation method, controls including counterbalancing, and one honest weakness |
| Observation | Overt or covert, participant or non-participant, structured or unstructured, the exact wording of each behavioural category as written before observing, the sampling rule and interval, whether two observers coded the same segment and how closely they agreed |
| Survey | Construct operationalisation, item and scale structure, an example item in its exact wording, administration and anonymity procedure, response rate and sampling method |
| Interview | Interview type, the interview schedule, the neutral probes you actually used, how it was recorded, and the participant's own phrasing rather than your summary of it |
That last one matters more than it looks. "I just do the easy thing first so it feels like I started" is evidence. Your paraphrase of it is not.
Add one honest weakness to every record. The extraneous variable you could not control. The category that turned out to be too vague. The item that was leading or double-barrelled. Those weaknesses are what 1b and Section B are built to reward, and they are impossible to reconstruct from memory.
The reflection that turns a practical into exam material
Immediately after data collection, answer four questions in writing.
Did anything vary that you did not intend to vary. Did any participant work out what was being tested, and how could you tell. Which items or categories did people hesitate over, because that hesitation is measurement data. And if you ran it again tomorrow, what one thing would you change first.
Then run each of the six concepts across the practical. Not in general terms, but specifically. Where could systematic distortion have entered, in the participant, the researcher, the sample, the wording or the coding, and in which direction would it push. What can this method legitimately say about cause. Is change actually being measured. What construct was claimed and what was actually measured. Whose perspective is visible and whose is missing. What ethical decisions protected participants, and how did those decisions affect the research.
Six lines per practical. Written on the day. That is the whole preparation for question 1b.
One ethics check worth doing now
For each of your four practicals, name the one ethical issue that mattered most and the specific step you took. Not a general principle. The actual step.
If you cannot do that for all four, those are the practicals to go back to first.
If you want a full record template plus worked models for each of the four methods, that is what the PsychExcel Paper 2 Student Guide is built around.