Sampling occurs when researchers select a suitable sample, which they intend to represent the ideas or characteristics of the whole population. This is done because it is often more cost and time efficient to gather primary data or evidence for a research hypothesis or question by selecting just a sample group of people rather than trying to gather data from the whole population.
Simple random sampling means that each individual within the population has an equal chance of inclusion in the sample.
For example, if you want to select 10 people randomly from a population of 100, you could write or print their names on separate pieces of paper, fold them up, mix them thoroughly and then select 10 pieces of paper. In this case, every person has an equal chance of being picked.
A systematic sample is obtained by selecting one person on a random basis, and then choosing additional people at evenly spaced intervals until the desired number of units has been obtained.
For example, say you have a list of 100 students, in alphabetical order, and you want to select a sample of 20.
Using systematic sampling, you divide 100 by 20, to get 5.
You then randomly select any number between 1 and 5. Suppose that the number you pick is 4: that will be your starting number.
You then select student number 4, and then select every 5th name on the list until you reach the end of the list. You will end up with 20 students in your sample
Cluster sampling involves taking random samples from clusters or groups that exist in the population.
A convenience sample is obtained by randomly selecting people from the population who are easy to access.
For example, you might interview the first 10 people you meet in one day, or randomly select 10 of your close friends to survey.
This can save time, money and effort, but it is the poorest way of selecting a sample as it can result in a lack of relevant data being collected or unreliable research outcomes.
To use this method, the researcher first needs to break the population into groups or divisions based on one particular characteristic or feature, such as education, income level, age or gender.
For example:
Group A = people who have completed less than five years of education
Group B = people who have completed five to 10 years of education
Group C = people who have completed 10 to 15 years of education.
A stratified sample is then obtained by selecting a simple random sample from each of these groups. The simple random sample can also be based on a proportion of the people in each group. For example, if Group A comprises 100 people, Group B only comprises 50 and Group C comprises 30, you might decide to select a 10 per cent sample from each group, meaning you would select 10 people from Group A, five from Group B and three from Group C.
Snowballing is a sampling technique where you find a participant and then they find other participants based on people that they know. For example, you may know someone who has been adopted and they agree to participate in your study. They then pass on details of other people they know who have been adopted.
The sample group is the representative sample that has been chosen to be the target population.
In order to determine the most appropriate sample size, you first need to consider the resources you have available, such as:
Money: how much will it cost to collect data from each person? This might include the costs of photocopying a questionnaire for each person, travelling to interview each person, and so on
Time: how much can you do within the available time? How many people do you have time to survey? How long will it take to collate the data you collect, based on the size of your sample?
Knowledge: do you know enough people who fit the population that you intend to study? For example, if you want to study the refugee population, how many refugees or refugee organisations do you know of?
Access to services: are you able to access the sample group? For example, if you want to research a topic specifically related to hospitals, how many hospitals do you have access to?
The sample size is how many representatives will make p the sample group and may be dependant on the focus of the research, the purpose, sampling method, time, money and access to services.