Quantitative researchers are often keen to examine characteristics of populations or make generalisations about group larger than their study samples.
For instance, a researcher would want to study features of home-makers in Argentina. Here collecting information from all the home-makers in Argentina is impossible. Therefore, the researcher would choose individuals and collect relevant from them. This is nothing but a sampling process. Sampling is widely used to research the entire population to gather actionable insights.
Quantitative research involves two types of sampling techniques: Probability and non-probability. Although many researchers rely mostly on probability sampling, non-probability sampling approach comes to an aid when there is a lack of access to the list of the population being studied.
Here’s a detailed look at probability and non-probability sampling
- Probability sampling
Probability sampling method uses a randomisation technique while choosing the units from the sampling frame to be included in the study sample. In this sampling technique, every member of the population has a known non-zero probability of selection.
For example, if you have a population of 50 people, every member of the population would have odds of 1 in 50 for getting selected.
This method includes various types of technique, such as:
- Simple random sampling – Here, the primary units are selected in such a way that every single unit in the population has an equal chance of getting selected.
For example, if a company has 100 employees, and the management team decides to conduct an activity by picking chits, then there is an equal probability of all 100 employees getting selected.
- Systematic sampling – In this approach, the members are selected at regular intervals of the population.
For instance, if you intend to collect sample of 50 people from a population of 100, then the population will be numbered from 1-100, and every 2nd individual (100/50) will be selected to be a part of the sample.
- Stratified sampling – In this method, the population is segregated into small groups such that they do not overlap and represent the entire population.
For e.g., if the research is being conducted on a set of people of different incomes, then people will be categorised as per the range of the income and conclusions from people belonging to different groups will be drawn.
- Cluster sampling – In approach involves the division of the entire population into sections or clusters.
For instance, if you are conducting research to determine the attributes of the refugees, then the refugees will be segregated as per their countries, and their attributes will be studied.
The benefit of probability sampling method is that sampling error can be calculated and guarantees a randomised selection process without bias.
- Non-probability sampling
Non-probability sampling method depends on the subjective judgement (involving a combination of theory & experience) of the researcher while choosing units from the population to be included in the sample. In this type of sampling, members of the population do not have an equal probability of getting selected. Therefore, it is not safe to consider that the sample fully represents the target population.
This approach involves different techniques, such as:
- Convenience sampling – In this method, the subjects are selected due to their convenient accessibility and proximity.
For example, to obtain reviews on buying experience, a psychology researcher can randomly interview individuals in a mall and acquire the necessary information.
- Judgemental sampling – Also, known as purposive sampling, this approach involves selection of units to be sampled on the basis of the researcher’s knowledge and judgement ability.
For e.g., if you want to identify individuals who are interested in pursuing a PhD, then upon obtaining the negative answer from the participant, the individual will be excluded from the sample.
- Snowball sampling – Here the additional participants are selected with the help of the initial participants.
For instance, surveying illegal immigrants is challenging. In such cases, you can interview a few of them can obtain results derived from their responses.
- Quota sampling – This approach involves the formation of sample involving individuals representing a population and selected according to qualities.
For example, if you want to determine the academic performance of high school students in relationship with the gender, then the subgroups must have the same number of participants of a particular gender.
Difference between probability and non-probability sampling technique
| Features | Probability sampling | Non-probability sampling |
| Population selection | Population is selected randomly | Population is selected arbitrarily |
| Selection criteria | Research design defines the selection criteria | Neither the selection criteria or the sample is defined |
| Hypothesis | There is an underlying hypothesis | Hypothesis is derived after conducting the study |
| Results | Results are unbiased | Results are biased |
Sampling is one of the key factors that determines the accuracy of your research outcome. Therefore, identify the type of sampling technique your study requires and collect samples precisely.