Sampling Methods (College Board AP® Psychology): Study Guide

Population, sample & representativeness

• At the start of the research process, the researcher must identify the target population

  • This is the full group of people the researcher is interested in studying and to whom they want to generalize their findings, e.g.

    • college students enrolled in introductory psychology courses

    • adults aged 40–60 diagnosed with clinical depression

    • first-generation immigrants living in urban areas

• Because it is rarely possible to study an entire population, the researcher selects a sample

  • A sample is a subset of individuals drawn from the target population to participate in the study

• The researcher then uses the findings from the sample to draw conclusions about the wider target population — this is called generalization

• The extent to which findings can be generalized depends on how representative the sample is of the target population

  • A representative sample accurately reflects the key characteristics of the target population (e.g. age, gender, ethnicity, socioeconomic status)

  • A sample that does not reflect these characteristics limits the researcher's ability to generalize findings beyond the sample itself

• Sampling bias occurs when the sample is not representative of the target population

  • This threatens the external validity of the study, e.g. a study conducted entirely on white, male, college-aged participants cannot be confidently generalized to the broader population

• Sampling bias can distort research outcomes and lead to conclusions that do not apply to underrepresented groups

Random sampling

• Random sampling involves selecting participants from the target population entirely by chance

  • This is so that every member of the population has an equal probability of being selected

• Methods of random sampling include:

  • writing the names of every member of the population on separate slips of paper, placing them in a box, and drawing names until the required sample size is reached

  • using random number generator software — particularly useful for large populations

• Random sampling is considered the least biased of all sampling methods because the researcher has no control over who is selected

Evaluation of random sampling

Strengths

• This sampling method eliminates researcher bias as the researcher cannot consciously or unconsciously favor certain individuals

• Random sampling produces a representative sample, which means findings are more likely to be generalizable to the target population

  • This strengthens the external validity of the study

Limitations

• Random sampling can be time-consuming and impractical

  • It requires access to a complete list of the target population, which is not always possible

• Not everyone selected may be willing to participate, which can reduce the sample size and introduce bias

  • Those who agree to take part may differ in important ways from those who decline

• Despite being random, there is no guarantee the sample will be representative

  • By chance, the sample could be unbalanced (e.g. an all-male sample could be randomly selected from a mixed population)

Convenience sampling

• Convenience sampling , also known as opportunity sampling, involves selecting participants who are available and willing to take part at the time the study is conducted, e.g.

  • approaching people in a shopping mall

  • recruiting college students from an introductory psychology class

  • selecting participants from those present at a community event

• Convenience sampling is the most commonly used sampling method in psychological research due to its practicality

Evaluation of opportunity sampling

Strengths

• Convenience sampling is quick, easy, and cost-effective as participants are readily available

  • This method is therefore practical for researchers working with limited time or funding

• As people have been approached and agreed to take part, this is more likely to mean that the research process runs smoothly

  • Having unwilling or resistant participants could damage the validity of the findings

Limitations

• Convenience samples are rarely representative of the target population

  • They only reflect those who happened to be available and willing to participate at a specific time and place, which limits generalizability

• The researcher may be prone to unconscious bias when approaching potential participants, e.g.

  • they may approach people they feel comfortable with

  • they may avoid certain social groups that they are wary of, e.g., males aged 18-25

  • This introduces sampling bias, which threatens the external validity of the findings

Stratified sampling

• Stratified sampling involves dividing the target population into subgroups, or strata, based on key characteristics relevant to the research, e.g.

  • age

  • gender

  • ethnicity

  • socioeconomic status

  • education level

• Participants are then randomly selected from each subgroup in proportion to their representation in the target population

  • E.g. if 30% of the target population is Hispanic, then 30% of the sample should be Hispanic

• This produces a small-scale reproduction of the target population within the sample

Evaluation of stratified sampling

Strengths

• Stratified sampling produces a highly representative sample

  • Because each subgroup is proportionally represented, the findings are more generalizable to the target population

• Stratified sampling reduces sampling bias by ensuring that no subgroup is over- or underrepresented in the sample

• Stratified sampling means that the researcher has control over which characteristics are used to create the strata

  • This allows them to prioritize variables that are most relevant to the research aim

Limitations

• Stratified sampling can be difficult to implement, as it requires detailed knowledge of the target population's composition, which is not always accessible

• Classifying every member of the population into a subgroup can be challenging

  • This is particularly the case for characteristics that are not clearly defined, e.g. socioeconomic status, ethnicity

• The process of identifying and recruiting participants from each stratum can be time-consuming and resource-intensive