Sampling Methods

When conducting research, psychologists cannot usually study everyone in the population of interest. Instead, they select a sample — a smaller group of people who are intended to be representative of the wider population. The method used to select this sample can significantly affect the quality and generalisability of the findings.

---

Key Terms

Term	Definition
Population	The entire group of people that the researcher is interested in (e.g. all GCSE students in the UK)
Sample	The smaller group of people selected from the population to take part in the study
Sampling	The process of selecting participants from the population
Generalisability	The extent to which findings from the sample can be applied to the wider population
Bias	A systematic distortion in the sample that means it does not represent the population accurately

flowchart TD
    A[Sampling Methods] --> B[Random]
    A --> C[Opportunity]
    A --> D[Stratified]
    A --> E[Systematic]
    B --> B1["Equal chance for all<br/>low bias, hard to do"]
    C --> C1["Whoever is available<br/>quick, biased"]
    D --> D1["Subgroups in proportion<br/>representative"]
    E --> E1["Every nth person<br/>objective, hidden patterns risk"]

---

Sampling Methods

1. Random Sampling

Every member of the target population has an equal chance of being selected.

How: Names are drawn from a hat, or a random number generator is used to select participants from a complete list of the population.

Advantage	Disadvantage
Unbiased — every person has an equal chance	Requires a complete list of the target population, which is often difficult to obtain
Most likely to produce a representative sample	Time-consuming and logistically difficult
Results are more generalisable	Selected participants may decline to take part, creating a biased sample

2. Opportunity Sampling

The researcher selects whoever is available and willing to take part at the time and place of the study.

How: The researcher approaches people in a convenient location (e.g. on campus, in a shopping centre).

Advantage	Disadvantage
Quick and easy — convenient for the researcher	Biased — only includes people who are available in that location at that time
Inexpensive	Not representative — may over-represent certain groups (e.g. students)
	Difficult to generalise to the wider population

3. Stratified Sampling

The population is divided into subgroups (strata) based on relevant characteristics (e.g. age, gender, ethnicity), and participants are randomly selected from each subgroup in proportion to their occurrence in the population.

How: If the population is 60% female and 40% male, the sample should also be 60% female and 40% male. Participants within each stratum are selected randomly.

Advantage	Disadvantage
Produces a representative sample that reflects the composition of the population	Time-consuming — requires detailed knowledge of the population's composition
Ensures that minority groups are included	Requires a complete list of the population and information about subgroups
Results are more generalisable	Difficult to implement in practice

4. Systematic Sampling

Participants are selected according to a fixed pattern from a list (e.g. every 5th, 10th, or 20th person on a list).

How: Starting from a random point on the list, every nth person is selected.

Advantage	Disadvantage
Objective — follows a fixed rule, reducing researcher bias	The list may have a hidden pattern that biases the sample
Easy to implement once the list is available	Requires an ordered list of the population
More structured than opportunity sampling	Not truly random — some members of the population may never be selected

---

Comparing Sampling Methods

Method	Representative?	Ease of Use	Bias Risk
Random	High	Low (difficult)	Low
Opportunity	Low	High (easy)	High
Stratified	High	Low (difficult)	Low
Systematic	Moderate	Moderate	Moderate

---

Generalisability

A sample is generalisable if the findings from the sample can be reasonably applied to the wider population. Generalisability depends on:

• Sample size — larger samples are generally more representative
• Sampling method — random and stratified sampling produce more representative samples
• Demographic characteristics — the sample should reflect the diversity of the population (age, gender, ethnicity, socioeconomic status)

Exam Tip: When evaluating a study, always consider the sampling method used. Opportunity samples (which are very common in psychology) may be biased and limit generalisability. This is a valid criticism of many studies.

---

Key Points

• A sample is a smaller group selected from the target population.
• Four sampling methods: random, opportunity, stratified, systematic.
• Random sampling is unbiased but difficult; opportunity sampling is easy but biased.
• Stratified sampling ensures proportional representation of subgroups.
• Generalisability depends on sample size, method, and representativeness.

---

The Real Cost of Biased Samples

A famous illustration of sampling bias is the 1936 Literary Digest poll of US voters, which predicted a landslide victory for one candidate based on a huge sample. The prediction was wrong because the sample was drawn from telephone owners and magazine subscribers — both wealthier than the average American at the time. The sample was large but biased, and the magazine's reputation never recovered. In psychology, biased samples rarely cause public scandals, but they routinely lead to findings that fail to generalise — which is one reason the field has faced a "replication crisis" in recent years.

The takeaway: a small, carefully drawn sample is often more valuable than a large, biased one.

---

Worked Practice: Sampling in Quick Scenarios

Identify the sampling method in each of the following.

Scenario A: A researcher takes every 10th name from a school register. → Systematic sampling.

Scenario B: A researcher puts all names from a list into a hat and draws 30. → Random sampling.

Scenario C: A researcher stands outside a coffee shop and asks the first 40 customers who agree. → Opportunity sampling.

Scenario D: A researcher ensures her sample is 60% female and 40% male (matching the college's population), with names randomly chosen within each group. → Stratified sampling.