5. sampling strategies in psychology

define population

the set of individuals with the particular characteristics under investigation in a research question

why can we not typically study the entire population

• access - may not have contact information about every individual in the population

• resources - may not have money or personnel to conduct a study with everyone in the population

• time - conducting a study with the entire population would usually take a long time

solution = get a sample = select some individuals from the population to take part

from population to sampling strategy

1. identify the population - balance breadth with precision

2. identify the relevant population characteristics that need to be properly represented in the sample

3. choose appropriate sampling strategy - maximise representativeness

sampling error

likely discrepancy between the results found in a particular sample and the results that would have been obtained in the population

amount of sampling error is influenced by two factors:

• sample size

• sampling strategy

size of sample

• larger sample → lower sampling error

• larger samples are more likely to be representative

sampling strategy

• approach taken to identify and recruit individuals from the population to take part in the study

• randomly chosen sample → lower sampling error

random sampling

selection at random from entire population

• likely to produce representative samples, but involve higher cost to identify and recruit eligible participants

stratified sampling

selection at random from sub-groups in the population

• likely to produce representative samples, but involve higher cost to identify and recruit eligible participants

opportunity/convenience sampling

self-selection in response to advertisement/call for participants made in a specific context or period

• May produce a representative sample

• Possible selection bias based on individuals’ presence in the recruitment context/period

pros and cons - random and stratified

• most likely to produce representative sample

• higher costs involved

pros and cons - opportunity/convenience

• lower cost to advertise study and recruit participants

• may not produce representative sample due to possible selection bias