Sampling (AQA A-Level Psychology): Revision Notes
Sampling
What is sampling?
Sampling refers to the process of selecting participants from a larger group to take part in research studies. A population includes everyone who could potentially be studied (for example, all teenagers in the UK), while a sample is the smaller group actually selected for the study.
The key goal of sampling is to create a representative sample - one that accurately reflects the characteristics of the target population. This allows researchers to generalise their findings from the sample back to the wider population. However, bias can occur when certain types of people are more likely to be selected than others, making the sample unrepresentative.
The quality of sampling directly impacts the validity of research findings. A biassed sample can lead to conclusions that don't apply to the broader population, limiting the usefulness of the research.
Psychologists use several different sampling techniques, each with distinct advantages and limitations.
Random sampling
Random sampling ensures that every member of the target population has an equal chance of being selected for the study. This can be achieved by placing all names from the target population into a container and drawing out the required sample size, or by using computer programmes to generate random participant lists.
Strengths of random sampling
- Unbiased selection: Since selection occurs without prejudice, there's no systematic bias favouring particular types of participants. This increases the likelihood of obtaining a representative sample.
- Strong generalisation: Because the sample should fairly represent the target population, research findings can be confidently applied to the broader group being studied.
Weaknesses of random sampling
- Impractical implementation: True random sampling proves difficult to achieve in practice. Researchers may struggle to obtain complete lists of target populations, and not all selected participants may be available or willing to participate.
- No guarantee of representativeness: Even with unbiased selection, random chance might still produce an unrepresentative sample. For instance, a random selection could theoretically choose only female participants, making findings non-generalisable to males.
Critical Point: True random sampling is often more theoretical than practical. Most studies claiming to use random sampling actually use modified versions due to real-world constraints.
Opportunity sampling
Opportunity sampling involves recruiting participants who are readily available and willing to take part in research. This might include approaching people in shopping centres, asking students in a school corridor, or recruiting friends and family members.
Research by Sears (1986) found that 75% of university psychology studies used undergraduate students as participants, largely due to their convenient availability.
Strengths of opportunity sampling
- Ease of formation: Creating opportunity samples requires minimal effort from researchers, as they simply approach people who are already present and accessible.
- Suitable for natural experiments: When studying naturally occurring events, researchers often have no control over participant recruitment, making opportunity sampling the most practical option.
Weaknesses of opportunity sampling
- Unrepresentative samples: Opportunity samples tend to exclude certain types of participants, creating systematic bias. For example, collecting data in a town centre during weekday working hours would exclude people at work or college.
- Self-selection bias: Participants choose whether to take part, and this decision itself may be influenced by personality characteristics or attitudes that affect the research outcomes.
Despite its limitations, opportunity sampling remains the most commonly used method in psychological research due to its practicality and cost-effectiveness.
Volunteer (self-selected) sampling
Volunteer sampling occurs when people actively choose to participate in research, often responding to advertisements, online posts, or recruitment notices. Participants essentially select themselves rather than being chosen by researchers.
Strengths of volunteer sampling
- Easy sample formation: Researchers need only create advertisements or notices, then wait for volunteers to come forwards, requiring minimal active recruitment effort.
- Reduced 'screw you' effect: Since participants volunteer willingly, they're less likely to deliberately sabotage the study through uncooperative behaviour or deliberate incorrect responses.
Weaknesses of volunteer sampling
- Highly unrepresentative: Volunteers tend to share certain characteristics - they may be more outgoing, confident, or interested in psychology. This creates samples that don't represent the general population.
- Demand characteristics: Eager volunteers may try harder to please researchers or provide responses they believe are expected, potentially distorting results rather than showing natural behaviour.
Beware of Self-Selection Bias: Volunteers often share similar personality traits (extraversion, openness to experience), which can significantly skew research results and limit generalisability.
Systematic sampling
Systematic sampling involves selecting every nth person from a complete list of the target population. Researchers first calculate the required interval by dividing the population size by the desired sample size, then select participants at regular intervals.
Worked Example: Systematic Sampling Calculation
To select 20 participants from a company's 1,000 employees:
Step 1: Calculate the sampling interval
Step 2: Select every 50th name from the employee list Starting with a random number between 1-50, then select every 50th person thereafter.
Strengths of systematic sampling
- Unbiased selection: The mathematical approach eliminates researcher bias in participant selection, increasing chances of obtaining a representative sample.
- Good generalisation: Results should be representative of the population unless specific characteristics repeat at regular intervals throughout the population list.
Weaknesses of systematic sampling
- Periodic traits: If the population list contains hidden patterns that coincide with the sampling interval, the technique becomes neither random nor representative. For example, if every fifth property on a street happens to be occupied by young people, selecting every fifth resident would create an unrepresentative age sample.
- Still not guaranteed representative: Mathematical selection doesn't guarantee that the sample will accurately reflect all population characteristics.
Stratified sampling
Stratified sampling creates a small-scale reproduction of the target population by maintaining the same proportions of different subgroups. Researchers first identify key characteristics relevant to their study (such as age, gender, or social class), then randomly select participants from each subgroup in proportion to their representation in the overall population.
For instance, if 12% of the target population falls between 20-30 years old, then 12% of the sample should be randomly selected from this age group. This ensures proportional representation across all relevant characteristics.
Strengths of stratified sampling
- Highly representative: By ensuring proportional representation of key subgroups, stratified samples closely mirror the target population's composition.
- Unbiased within strata: Random sampling within each subgroup eliminates selection bias while maintaining representativeness.
Weaknesses of stratified sampling
- Requires detailed population knowledge: Researchers must possess comprehensive information about population characteristics, which may not always be available or accessible.
- Time-consuming process: Dividing populations into appropriate strata and then conducting random sampling within each category requires considerable time and planning effort.
Key Points to Remember:
- Random sampling gives everyone equal selection chances but can be impractical to implement properly
- Opportunity sampling is convenient and quick but often produces biassed, unrepresentative samples
- Volunteer sampling attracts willing participants but tends to recruit people with similar characteristics
- Systematic sampling uses mathematical selection to reduce bias but can encounter problems with periodic population patterns
- Stratified sampling creates the most representative samples but requires detailed population knowledge and significant preparation time