In this post
One important consideration for researchers is who they are going to carry out their research on. Participants who take part in research are known as a ‘sample’, and these people are a smaller group that is drawn from a ‘target population’. A target population is all of the people that a researcher wants to study.
It is best if a sample is representative of the target population, which means that the people in the same group provide a good cross-section who have a range of characteristics which are relevant to the study. Sometimes, it is clear that samples are not representative. For example, if a study wanted to find out if children are able to recognise ten different types of animal by the age of two and the sample was made up of all girls, then this would not be representative at all.
Sometimes, however, researchers want to narrow down their target population and therefore it is acceptable if they only research one sex or one culture, as this is linked to what they want to find out. For example, if they wanted to find out if British mothers use certain types of language when communicating with their babies then it would be acceptable for the researcher to only use British mothers in their sample as this is their target population.
There are a number of techniques that researchers can use to try and ensure that the sample they use is representative, and the four most common types of sampling are as follows:
- Random
- Stratified
- Volunteer
- Opportunity
Random sampling
A random sample means that the researcher has drawn participants out by chance – like pulling names from a hat. With this technique, it is thought that anyone has an equal chance of being selected and therefore there is likely to be less bias and more chance of a good cross-section of characteristics.
Researchers have to come up with a way of randomly selecting their participants. So, for example, if they conducted research in a school, they could choose every third name on the register.
Stratified sampling
The researcher identifies the different types of people that make up the target population and works out the proportions needed for the sample to be representative.
A list needs to be made of each variable and the sample must be representative of these variables. For example, if a researcher wanted to find out if drinking a specific type of alcohol could predict the level of someone’s educational achievement within a workplace then the researcher would have to work out the relative percentage of each type of alcohol drinker – wine, vodka, bourbon, gin, beer etc. So, it may be that wine drinkers accounted for 40% of the sample, 20% for vodka, 15% for bourbon, 15% for beer and 10% for gin, then the sample must then contain all these groups in the same proportion as in the target population – i.e. the workplace.
Volunteer sampling
Exactly as it sounds, volunteer sampling consists of participants who have volunteered themselves to take part in a study. Researchers can advertise for people to take part in a study and people put themselves forwards to take part – in other words, they ‘self-select’. Researchers can attain this kind of sample by newspaper, internet or magazine adverts or by putting up posters where people who may be interested in the study will see them, such as in the halls of residence of a university.
Opportunity sampling
This is a very straightforward method of obtaining a sample of participants, as the researcher simply selects people who are convenient and willing to take part – such as asking people as they pass by in the street. Researchers often use psychology students, as they are readily available and keen to take part, especially if a payment is offered in exchange for their participation.
Opportunity sampling may be summed up as using someone who was in the right place at the right time and was willing to take part.
| Type of sampling method | Strengths of this method | Weaknesses of this method |
|---|---|---|
| Random | Very representative of the target population. Should eliminate bias within the sample. | Difficult to achieve and is costly in terms of both time and money. There is no control over who is chosen and so a ‘freak’ sample could be chosen, e.g. everyone with a really bad memory for a memory test. |
| Stratified | The sample should be very representative of the target population, allowing results to be generalised. | Extremely time consuming and almost never used in psychological research. |
| Volunteer | Easy to apply. Participants will be very willing to take part. | Bias is very common as individuals are interested in what is being researched. Sample will almost certainly not be truly representative. |
| Opportunity | Quick and easy to apply. | May not be representative as all may be one type of person, such as a psychology student. Sample will almost certainly not be truly representative. |
