**Tips, Types and Techniques for Sampling in (Marketing) Research**

In statistical analysis, a

**sample** is a subset of individuals or objects from a larger population.

**Sampling** is the process used in statistical analysis for selecting a predetermined number of observations (a sample) from a larger population or market that you want to collect data from in your research. For example, if you are researching the opinions of European managers about decision-making, you could survey a sample of 100 European managers. In statistics, sampling allows you to test a hypothesis about the characteristics of a larger, total population. In other words, sampling gives us a gist of the whole population or market.

In all of this it is clear that the quality of the results of your (market) research are going to depend on how representative or

significant the sample was as compared to the whole population or market. Unless you are extremely lucky.

Sampling can be broadly classified into 2 types which are Probability Sampling and Non-Probability Sampling.

In

**Probability Sampling**, random sampling techniques are used and there is an equal chance of everyone being chosen to the sample. It is mainly used in quantitative research. If you want to produce results that are representative of the whole population, you need to use a probability sampling technique.

In

**Non-Probability Sampling** certain predefined factors are being used which will result in not everyone having the same chance to participate in the sample. This type of sampling is easier and cheaper to access, but it has a higher risk of sampling bias, and you can't use it to make valid statistical inferences about the whole population. That's why Non-probability sampling techniques are often appropriate for exploratory and qualitative research. In these types of research, the aim is not to test a hypothesis about a broad population, but to develop an initial understanding of a small or under-researched population.

The two main types of sampling have various subtypes which are the main methods of sampling. The figure below gives you an overview:

**Probability Sampling Techniques are**:

- SIMPLE RANDOM SAMPLING: This is the simplest way of sampling which takes very little time because it is done randomly. Here everyone has the same probability of being chosen.

*For example*: Suppose a manager has to choose 6 people for a project then she might choose any 6 employees out of the workforce by taking chits out of a bowl or simply using the simple "randbetween" function in Excel of their social security numbers.
- CLUSTER SAMPLING: In this type of sampling, the entire population is divided into small clusters or sections which represent the population. Here the clusters should be mutually homogeneous and internally heterogeneous. In other words, each subgroup should have similar characteristics as the whole population. After that out of these, the sample population is selected by simple random sampling.

*For example*: If someone has to determine the consumption of a type of product in a country, then it may be easier to divide the country into clusters of regions and then proceed with simple random sampling.
- SYSTEMATIC SAMPLING: As the name suggests it's a type of sampling where some system or pattern is involved. So if a researcher has to select a sample of 100 people out of 1.000 then he might number all subjects and take 100 people following some pattern. Because a predefined pattern is involved, this is a fast sampling process.

*For example*: The researcher takes every 10th person from the population for the sample, like selecting all those that have as last digit a 7.
- STRATIFIED RANDOM SAMPLING: Here the entire population is divided into different strata/groups based on the relevant characteristic (e.g. gender, age range, income bracket, job role) that represent the entire population. It is different from cluster sampling as in stratified random sampling the groups are made based on certain qualities.

*For example*: Suppose people are divided based on age being below 18, 18-30, 30-50, and 50 and above, then the researcher can take proportional amounts of people from each group. Suppose the proportion of people between 18-30 is 30% of the total population then the researcher will select 30% of the people in the overall sample from that group.

**Non Probability Sampling Techniques are**:

- CONVENIENCE SAMPLING: As the name suggests, this method is based on what is the most convenient way for the researcher to take the sample. The researcher could conduct a survey with his students if he/she happens to be a professor. This type of sampling is chosen when there are constraints of time and cost during the research.

*For example*: If a student is doing research then it's easy for him to access the people of the area he lives in so he selects people in his locality for the survey.
- QUOTA SAMPLING: In a quota sampling approach, a researcher sets some quota to certain subgroups and takes samples until the quota is fulfilled. Suppose for the research he needs a 30% population below 18 years of age then he will sample it in such a way. This allows a researcher to effectively sample a subgroup that is of more importance to the study.

*For example*: If a researcher wants to compare the behavior of teenagers with other age groups, he/she will have a certain quota for the percentage of a sample of teenagers versus the others.
- JUDGEMENT SAMPLING: Judgement sampling is based on the judgement or choice of the researcher. If a researcher wants to survey only the people with certain traits, then he/she will add some criteria to filter the sample.

*For example*: Suppose a researcher wants to understand the behavior of the people who are interested studying in the medical field then he will add a question in the survey such as "are you interested to study in medical field" then for the further research he will select only the people who have answered yes. He then assumes that those are the ones who are interested studying in the medical field (although they might have lied).
- SNOWBALL SAMPLING: Snowball sampling is used when it is quite difficult to find people needed for the sample. Using snowball theory, it becomes easier for the researchers to find such samples. Snowball theory suggests that once the researcher finds a subject, he asks them to recruit more subjects which they know. This is like a chain process where the subjects recruit more and more people and thus the sample gets bigger. Suppose you want to sample people having a rare disease then this type of sampling could be used.

*For example*: A researcher wishes to sample people having a particular form of Cancer, but it's quite difficult as this is a rare form and often people would not want to come forward publicly. So the researcher needs to find some people who have this form of cancer, so then asks them to find more subjects as they are more likely to meet such people when they go for chemotherapy sessions or other treatments.

I hope if you read this you react and share additional interesting things about sampling.

*Sources:*

Dan Fleetwood, "Types of Sampling in social research", Questionpro

"Probability Sampling", Statisticshowto

Nargundkar, R. (2017), "Marketing Research: Text and Cases", McGraw Hill Education, Third Edition.