Try a free Voxco Online sample survey!
SHARE THE ARTICLE ON
Sampling is used to appropriately select elements of a target population to create a sample group that is representative of the entire population. Researchers need sample groups to make inferences about a sample group that can be generalizable to the whole target population.
Researchers use different sampling methods depending on their resources, time limitations, research topic etc. Different methods of sampling are apt for different studies. In this article we will be discussing the types of sampling.
Download the guide to understand step-by-step process of selecting the best sample for your next survey research
There are two broad categories of sampling methods used for social research. They are as follows:
1. Probability sampling:
Methods of sampling under this category are based on the theory of probability. Probability sampling methods ensure that each element in the population has an equal and known chance of being represented in the sample group. For example, if I have a target population of 100 people, each person will have a 1/100 chance of being selected as a respondent in the study.
The following are the four main types of probability sampling methods:
2. Non-Probability Sampling:
Methods of sampling under this category, on the other hand, do not give all respondents an equal chance of being selected in the sample group. Non-probabilistic methods rely on judgment, convenience, and/or logic to select elements instead.For example, a researcher may choose to survey those people who are easily and conveniently available to them.
There are four main types of non-probability sampling methods:
This method of sampling is the easiest and most basic method of probability sampling. It uses the “lottery method” or “random number tables”, for example, to choose elements from a population. Each element is given a number and softwares/processes that give random outputs are used to pick the number of elements defined by the sample size.
For example, if my target population is the adult population in Las Vegas, then I must have a list of each element in this population. I can then use certain softwares, Excel for instance, to input every element in the list and use commands that pick a certain number (sample size) of participants to be selected in the sample group randomly.
2. Systematic Sampling
Systematic sampling is where a researcher selects an interval and random starting point in order to choose their sample. The fixed interval can be calculated by dividing the target population by the chosen sample size.
For example, if I’m conducting a study on students between grade 9-12 from XYZ school, I can use stratified sampling to select a sample group. Assuming there are 300 students in the target population, and the sample size is 10, the interval will be 30 (300 divided by 10). Then, I will pick a number between 1 and 30 (random starting point), after which I will pick every 30th element on my list until I have 10 students for my sample group.
3. Stratified Random Sampling
This is a method of probability sampling that involves dividing the population into subsets, or strata, based on shared characteristics. These subsets are mutually exclusive and collectively exhaustive, so as to eliminate the overlapping of elements in subgroups. The variables used to define these subsets can be age, occupation, vicinity, gender etc. After the subgroups of the population are defined, the researcher selects elements from each of these subsets using SRS. Being a crucial social research method, systematic sampling is used when a researcher wants to ensure certain groups of the population are properly represented in the study.
For example, if a study is trying to determine differences in spending habits of adults of different age groups, stratified sampling can be used to select the sample group. First, the population will need to be broken down into subgroups according to their age. Then SRS can be used to select elements from each of these strata.
4. Cluster Sampling
Cluster sampling is a method of probability sampling where populations are divided into clusters defined by predetermined variables. These clusters are mutually exclusive and collectively exhaustive, hence there is no overlap of elements in clusters. After these subpopulations are formed, certain clusters are then eliminated to narrow down the population before SRS or stratified random sampling is used to select elements. The predetermined variable in cluster sampling is usually geographical area.
For example, if I’m conducting a study across the United States, I can consider each city to be a cluster/subpopulation in my target population. To narrow down this population, I will eliminate certain clusters (or cities, in this case) before I use SRS to select elements from the narrowed down American population.
1. Quota sampling
Quota sampling uses “control characteristics” to categories a target population into multiple subpopulations with shared characteristics. After these subgroups are defined, the researcher chooses elements from each subgroup using non-probability sampling techniques such as convenience or judgment. This method of sampling is similar to stratified random sampling as both these methods divide the population into subgroups based on certain variables. However, the main difference between the two is that in stratified random sampling SRS is used to select elements from the subgroups whereas in quota sampling, judgment or convenience is used instead.
For example, if the participation of respondents from every city in Canada is critical to a study, then the researcher must group participants city wise and choose elements from each of these subpopulations using convenience or judgment.
2. Snowball sampling
Snowball sampling is a method of non-probability sampling where the researcher uses their initial group of participants to help create and identify a larger network of those who qualify to be a part of the target population. This method of sampling is often used when the target population of a study is really small, hard to find, and/or inaccessible.
For example, in a study about homeless people, a researcher may ask homeless people that are readily available to them to give a list of areas where more homeless people can be found. In this case, the researcher is using one element, or a few elements, of the target population as a resource to access more people in that population.
3. Judgmental sampling
Judgmental sampling, also known as purposive sampling, is a quick, low-cost method of non-probability sampling. In this method, the researcher uses their judgment, logic, and expertise to select participants to be a part of the sample.
For example, if a survey’s target population is marketing experts, then a researcher may choose to interview any marketing experts they come across.
4. Convenience Sampling
Convenience sampling, also known as accidental sampling, is a non-probability sampling process carried out at a researcher’s convenience. This means that the researcher chooses respondents whenever and wherever they are met. This method of sampling is used when there is a limitation of time or if certain elements of the population are not easily come across.
For example, if I want to study the buying behaviors of sporting good customers, I may visit different sporting good stores in my city in order to survey different customers at these stores. These customers will be a part of my sample group.
The Art of Negotiating with a Customer you don’t want to Lose SHARE THE ARTICLE ON Share on facebook Share on twitter Share on linkedin
All you need to know about Customer Experience Measurement Tools Customer Experience Ensuring an excellent customer experience can be tricky but an effective guide can