Sampling is the process through which researchers strategically select participants for their studies so they can answer their research question. When writing a research question, most researchers will identify a ‘population’, or group of people, in whom they are interested. For example, if you wanted to understand the impact of COVID-19 lockdowns on the mental health of older people in the United Kingdom, you might define your population of interest as being ‘people aged 75+ living in the United Kingdom during the lockdown periods’.
Usually, quantitative researchers want to draw general conclusions about their population of interest. For example, your research related to the impact of COVID-19 lockdowns on the mental health of older people might reveal that the effect was negative or positive.
To arrive at such or similar conclusions, researchers will rarely be able to study every single member of their population of interest. Instead, they must identify a smaller group of the population which is ‘representative of’ the broader population on characteristics such as age, race and sex. This process is called ‘sampling’ and the selected group a ‘sample’.
There are two main quantitative sampling techniques:
- Probability sampling
- Non-probability sampling
In a probability sample, every member of the population of interest is equally likely to be chosen to become part of the sample. For example, to recruit a representative sample of members of a social support group, a researcher might randomly select 10% of all listed members. As the probability sample is selected by chance, it is assumed to be representative of all members of the support group in terms of age, gender, ethnicity, education, etc.
In a non-probability sample, some members of the population are more likely to become part of the sample than other members of the population. This might be because they react to an advertisement, they know someone else who took part in the study, or they were actively approached by the researcher. Therefore, a non-probability sample is considered to be less representative, because it is not selected by chance. This means that members of the sample may have certain characteristics (e.g., helpful, interested in research) or respond in a certain way (e.g., they only took part to make their opinion heard). Nevertheless, non-probability samples can be easier to obtain and can still produce meaningful insights for researchers.
(Author: Leonie Ader)