What you'll learn

• How to distinguish target populations, sampling frames and samples.
• The main participant sampling methods used in A-Level Psychology research.
• How to evaluate sampling methods using bias, representativeness, generalisability, practicality and ethics.
• How event sampling and time sampling work in observations.

Why participants matter

In psychology, researchers rarely study every person they are interested in. Instead, they study a smaller group and then judge how far the findings can apply more widely.

That makes participant selection very important. A study can have careful controls and accurate statistics, but if the sample is biased, the conclusions may not generalise well.

The process usually moves from a broad target population to a practical sampling frame, then to the final sample.

Sampling methods: the big divide

Sampling methods are often grouped into probability sampling and non-probability sampling.

Random sampling

Random sampling is a probability method where every person in the sampling frame has an equal chance of being selected. For example, a researcher could number every name on a school register and use a random number generator.

AO3: strengths and weaknesses

A strength is that random sampling reduces researcher bias because the researcher does not choose who looks useful or convenient. This can improve representativeness.

A weakness is that it requires a complete and accurate sampling frame. It can also be time-consuming, and selected people may refuse to take part, creating non-response bias.

Systematic sampling

Systematic sampling is a probability method where the researcher selects every nth person from an ordered sampling frame after choosing a starting point.

For example, if a researcher needs 20 people from a list of 200, they might select every 10th person.

AO3: strengths and weaknesses

Systematic sampling is quicker than pure random sampling and spreads participants across the sampling frame. However, it can be biased if the list has a hidden pattern. For example, choosing every 5th name from a seating plan could over-represent one part of the classroom.

Stratified sampling

Stratified sampling is a probability method where the target population is divided into important subgroups, called strata, and participants are randomly selected from each stratum in proportion to their presence in the target population.

Strata might include gender, age group, year group, ethnicity, diagnosis, or any variable relevant to the research question.

AO3: strengths and weaknesses

Stratified sampling can be highly representative because key subgroups are included in the correct proportions. This is useful when a variable may affect the behaviour being studied.

However, it takes more planning than opportunity sampling and requires accurate information about the population. It also only controls for the strata chosen; other variables may still be unevenly represented.

Opportunity sampling

Opportunity sampling means selecting people who are available and willing at the time. For example, a student researcher might ask classmates in the common room to complete a questionnaire.

AO3: strengths and weaknesses

A strength is practicality. It is quick, cheap and useful for classroom practical investigations.

A weakness is sampling bias. People who happen to be available may share characteristics, such as being from the same school, age group or friendship network. This limits generalisability.

Ethically, the researcher must avoid pressuring people just because they are nearby. Participants still need informed consent, the right to withdraw, confidentiality and debriefing.

Quota sampling

Quota sampling is a non-probability method where the researcher sets targets for certain categories, then recruits people until each quota is filled.

For example, a researcher may decide to recruit 10 male and 10 female participants, then approach people in a town centre until both quotas are complete.

AO3: strengths and weaknesses

Quota sampling can improve representativeness compared with pure opportunity sampling because important categories are deliberately included.

However, it is still non-random. The researcher may select the easiest or most approachable people within each category, causing interviewer bias. It also only represents the variables used for the quotas.

Self-selected sampling

Self-selected sampling, also called volunteer sampling, occurs when participants put themselves forward. This might happen through an advert, email, online post or poster.

AO3: strengths and weaknesses

A strength is that volunteers are likely to be motivated and have given consent clearly.

A weakness is volunteer bias. People who choose to take part may be more confident, more interested in psychology, or more affected by the topic than those who ignore the advert.

This matters when evaluating studies. For example, if a stress study attracts only highly stressed students, the sample may exaggerate stress levels in the wider population.

Snowball sampling

Snowball sampling involves asking initial participants to recruit further participants from their social networks.

This is especially useful for hard-to-reach groups, such as people with rare experiences, hidden populations, or groups where trust is important.

AO3: strengths and weaknesses

A strength is access. Snowball sampling can reach participants who would be difficult to find through public advertising.

A weakness is network bias. Participants may recruit people similar to themselves, so the sample may be narrow. There are also ethical issues: confidentiality must be protected, and potential participants should not feel pressured by friends, family or group members to take part.

Applying sampling to studies

Sampling is useful for AO3 evaluation of named studies. For example, Freud’s study of Little Hans (1909) used one child, so the sample was extremely small and unrepresentative. This limits generalisability, although the detailed qualitative data may still be valuable.

In Loftus and Palmer’s eyewitness testimony research (1974), students were used as participants. This made the research practical and controlled, but student samples may not represent all real eyewitnesses.

In Raine, Buchsbaum and LaCasse (1997), murderers pleading not guilty by reason of insanity were compared with matched controls. Matching helped control participant variables, but the sample was highly specific, so findings should not be overgeneralised to all offenders.

Observational sampling techniques

Participant sampling is about who takes part. Observational sampling is about when or which behaviours are recorded during an observation.

Event sampling

In event sampling, the researcher defines a behaviour clearly and records each occurrence. For example, in a playground observation, a researcher might record every time a child pushes another child.

Event sampling is useful for behaviours that are clear, countable and not happening constantly.

A weakness is that it can miss context or duration. Two aggressive incidents may be counted equally even if one lasts one second and another lasts one minute.

Time sampling

In time sampling, the researcher records behaviour at specific time intervals, such as every 30 seconds or every two minutes.

This is useful when behaviour is frequent or continuous, such as time spent on-task in a classroom.

A weakness is that behaviour between intervals may be missed. If a child is off-task for 20 seconds but back on-task when the interval is recorded, the observation may underestimate off-task behaviour.

Ethics when recruiting participants

Sampling is not just a methodological issue. It is also ethical.

Researchers should follow the BPS Code of Ethics and Conduct, including informed consent, avoiding unnecessary deception, right to withdraw, protection from harm, confidentiality and debriefing.

Recruitment adverts should be honest about what participation involves. Researchers should avoid coercion, especially when recruiting students, employees, patients, children or vulnerable groups.

If non-human animals are used, researchers must consider additional safeguards such as the 3Rs: replacement, reduction and refinement. This means avoiding animal use where possible, using the fewest animals needed, and minimising suffering.