Sampling affects who enters a study and how far findings can be generalized. In IB Psychology, evaluation means judging each technique by its usefulness, bias, practicality, and likely impact on research quality.

Why sampling matters

Psychologists usually cannot study every member of a target population, so they select a sample instead. The way that sample is chosen matters because it influences representativeness, the amount of bias in the study, and whether findings can be generalized beyond the participants actually tested.

When evaluating sampling, it is important to ask not only whether the method is convenient, but also whether it fits the research aim. Some techniques are fast and inexpensive, while others are stronger for producing a balanced sample.

Key criteria for evaluation

• Representativeness: Does the sample reflect the main characteristics of the target population?

• Bias: Does the method make some people more likely to be selected than others?

• Practicality: Is the method realistic in terms of time, cost, and access?

• Generalizability: Can the findings reasonably be applied to the wider population?

• Suitability: Is the technique appropriate for the specific group being studied?

Main sampling techniques

Self-selected sampling

In self-selected sampling, participants volunteer to take part after seeing an advertisement, online post, email invitation, or similar recruitment message. This method is common in surveys and internet-based research.

A major strength is that it is efficient. Researchers can reach many possible participants quickly, and volunteers have already shown willingness to participate, which may improve completion rates. It also supports informed consent because people actively choose to join.

A major limitation is volunteer bias.

People who volunteer may differ from those who do not. They may have stronger opinions, more free time, greater confidence, or more interest in psychology. This can reduce representativeness and weaken generalizability. Self-selected samples are therefore practical, but often less balanced than probability-based methods.

Opportunity sampling

In opportunity sampling, researchers choose participants who are available at the time and place of the study. This may include students in a classroom, people in a hallway, or individuals already present in a setting.

Its main advantage is speed. It is usually cheap, easy to organize, and useful when researchers have limited resources. For small-scale studies, this can make data collection possible when more complex methods are unrealistic.

However, opportunity samples are often narrow and biased toward accessible people. The sample may reflect one location, one time of day, or one social group rather than the wider population. Because of this, opportunity sampling tends to have low representativeness. It is easy to use, but often weak if the aim is broad generalization.

Random sampling

In random sampling, every member of the target population has an equal chance of being selected. This reduces researcher choice in participant selection and helps limit selection bias.

Random sampling usually requires a sampling frame. If the frame is accurate and selection is truly random, this method can produce a more representative sample than self-selected or opportunity methods. It is often seen as strong for quantitative research where generalization is important.

Its limitations are practical. A complete sampling frame may be difficult or impossible to obtain. The process may also take more time and organization than simpler techniques. In addition, a random sample is not automatically representative, especially if the sample is small or many selected people do not respond. Random sampling reduces bias, but it does not remove all sampling problems.

Stratified sampling

In stratified sampling, the target population is divided into relevant subgroups, called strata, such as age groups, gender categories, or other characteristics.

This diagram shows stratified sampling as a two-stage process: first partition the population into meaningful strata, then draw random samples within each stratum. Visually, it reinforces the core evaluation claim that stratification increases representativeness by preventing small-but-important subgroups from being missed, provided the strata are well-chosen. Source

Participants are then selected from each stratum, often in proportion to their presence in the population.

The main strength of stratified sampling is improved representativeness. Important subgroups are less likely to be missed, so the final sample may reflect the population more accurately than a simple random sample. This is especially useful when some groups are small but still important to include.

The main limitation is complexity. Researchers must know the composition of the population in advance and must choose appropriate strata. If the strata are poorly selected or population data are inaccurate, the method becomes weaker. Stratified sampling is often stronger than simple random sampling for balanced representation, but it is more demanding to carry out correctly.

Snowball sampling

In snowball sampling, existing participants recruit other participants from their own social networks.

This diagram visualizes snowball (chain-referral) sampling as a recruitment network that expands outward from initial “seed” participants. It helps explain why snowball sampling can efficiently reach hidden populations, while also illustrating how reliance on social ties can concentrate the sample within a few connected clusters (network bias). Source

The sample grows gradually through participant referrals.

This technique is especially useful for hard-to-reach or hidden populations, where no sampling frame exists and trust is important. It can help researchers access groups that are difficult to contact through ordinary methods.

Its main weakness is network bias. Participants often recruit people similar to themselves, which can make the sample unusually homogeneous. This reduces representativeness and can limit generalizability. Snowball sampling may also raise confidentiality concerns because recruitment depends on social connections. It is valuable for access, but usually weak for producing a broad cross-section of a population.

Comparing techniques in evaluation

No sampling technique is always best. The best choice depends on the research aim and the population being studied. If a study needs fast recruitment, opportunity or self-selected sampling may be justified. If the goal is stronger generalization, random or stratified sampling is usually better. If the group is hidden or difficult to identify, snowball sampling may be the most realistic option.

Common exam judgments

In IB answers, evaluation should go beyond naming strengths and weaknesses. Explain why a strength matters and how a limitation affects research quality. Strong responses use terms such as representativeness, bias, generalizability, sampling frame, and practicality accurately. They also compare methods rather than treating one technique as simply good or bad in all situations.