A sample is a subset of a population where analysing the population is either impossible or very difficult. In order to collect reliable data, extracting the right sample from the population is a must. “When we sample, we aim to make the sample as representative of the population as possible so that we can draw valid and reliable inferences about the population.“, Bocconi University.
How do we find it then? We can consider that each person in a population has the same probability to answer in a specific way. For example, if we choose a random person on the street, the probability of him/her choosing either A or B is equal, which is 50%. It is called Probability sample – the members of the sample are chosen on the basis of known probability (e.g. the same probability of being chosen).
However, in non-probability sample – we are not sure about the fact that the element chosen is representative of the population since it was chosen just because it was easy to obtain. For example, if we are conducting a survey on the preferences of Europeans, choosing a sample from Asia, just because it is easier to obtain is not logical because it is (refer to the lipstick photo) not a good representation.
It is very important to choose the relevant representative sample. For example, a survey about a lipstick would give more reliable answers if the possible users, women aged 20-60 (unless this is the case), are questioned.
The sample should have an opinion about the product or the brand to give a reliable answer.