We call “sampling” the statistical procedures used to select representative samples of the population to which they belong and which constitute the object of study of a given research.
In this article we will analyze the different types of sampling that exist, both random and non-systematic.
Sampling in inferential statistics
In statistics, the concept “sample” is used to designate any possible subset of a given population. So when we talk about a sample, we are referring to a certain set of topics that come from a larger group (the population).
Inferential statistics is the branch of this discipline which deals with study samples to make inferences about populations where these start from. Unlike descriptive statistics, the task is, as the name suggests, to describe in detail the characteristics of the sample, and therefore ideally of the population.
However, the process of statistical inference requires that the sample in question be representative of the reference population so that conclusions obtained can be generalized on a small scale. In order to promote this task, several have been developed sampling techniques, i.e. obtaining or selecting samples.
There are two main types of sampling: random or probability and non-random, also called “non-probability”. In turn, each of these two broad categories includes several sampling classes that differ on the basis of factors such as the characteristics of the reference population or the selection techniques employed.
Types of random or probability sampling
We speak of random sampling in cases where all subjects in a population have the same probability of being chosen as part of the sample. Samples in this class are more popular and useful than non-random samples, primarily because they have high representativeness and allow sample error to be calculated.
1. Simple random sampling
In this type of sampling, the relevant variables in the sample have the same probability function and are independent of each other. The population must be infinite or finite with reconstitution of the elements. Simple random sampling is most widely used in inferential statistics, But it is less effective in very large samples.
Stratified random sampling involves dividing the population into strata; an example of this would be to study the relation between the degree of vital satisfaction and the socio-economic level. A certain number of subjects are then extracted from each of the strata in order to maintain the proportion of the reference population.
3. From conglomerates
In inferential statistics conglomerates are sets of population elements, Like schools or public hospitals in a municipality. When performing this type of sampling, the population (in the examples, a specific locality) is divided into several conglomerates and some of them are chosen at random for study.
In this case, it begins by dividing the total number of subjects or observations that make up the population by what is to be used for the sample. Subsequently, a random number is chosen from the first and this same value is constantly added; the selected items will be part of the sample.
Non-random or non-probability sampling
Non-probability sampling uses criteria with a low level of systematization that aim to ensure that the sample has a certain degree of representativeness. This type of sampling is mainly used when it is not possible to make others of random type, Which is very common due to the high cost of control procedures.
1. Intentional, opinionated or of convenience
In intentional sampling, the researcher voluntarily chooses the elements that will constitute the sample, assuming that these will be representative of the reference population. An example that will be familiar to psychology students is the use of students as an opinion sample by university professors.
2. Snowball or chain sampling
In this type of sampling, researchers establish contacts with certain subjects; then they get new attendees for the show to finish it. Snowball sampling is generally used when working with hard-to-reach populations, As in the case of drug addicts or members of minority cultures.
3. Paid or accidental sampling
We speak of quota sampling when researchers choose a precise number of subjects who meet certain characteristics (for example, Spanish women over 65 with severe cognitive impairment) based on their knowledge of the strata of the population. Accidental sampling it is frequently used in surveys.