# 5 Common Sampling Techniques Used in Business Research

Topics: Sampling, Stratified sampling, Sampling error Pages: 11 (2604 words) Published: January 29, 2013
Question 1:
discuss any five (5) common sampling techniques used in business research. Support you answer with relevant examples.

Simple random sampling:
The simple random sampling is one of the most widely-used random sampling method. The term “random” here does not mean a haphazard selection as many people think. The “random” in this method means each member of the population has equal opportunities being chosen be subject and no one in the identified population who could not be selected in this method. For example, the teacher wants to choose 5 people in QTB class to stand up and introduce themselves. In order to perform random sampling, each member has to have a specific number as an ID, and those number are put in a random sample list or termed sampling frame. In the example, sampling frame would be the class list. The mechanical and primitive method would be the lottery method. Each number is placed in a bowl or a container and mixed thoroughly. After that, the researcher picks numbers tags from the container without any awareness of what numbers that are. All the individuals bearing the numbers picks are the subjects for the study. Thank to advance technology; another way to perform this sampling would be using computer or calculator to do a random selection from the population. There are two types: sampling without replacement and with replacement in simple random sampling. In the first example about choosing students to introduce themselves, the student who have talk about themselves could not be chosen one more time. So the teachers need to remove their number out of the sampling frame. When we look at another example like lottery, the numbers are picked, and then they are put back to the container. Those numbers, which are put back, may be selected more than once. Simple Random sampling has its advantages and disadvantages. On one hand, the best thing about this random sampling is that it is easy to perform. Moreover it is also considered as an unbiased random selection since every member is given equal chances of being selected. On the other hand, there is the most obvious limitation of simple random sampling method is its sampling frame required. The sampling frame must be complete and up to date, which is not usually available for large population.

Stratified random sampling:
Stratified random sampling, which is a variant of probability sampling technique, is used when population may have different value for the responses of interest. The researcher wants to highlight particular subgroups in the whole population. In this case, unlike the simple random sampling, we divide the population into groups that are called strata, than randomly selects the final subjects from the different strata. Each individual or unit in a stratum has same opportunity to be chosen. In order to give equal chance to each unit, the researcher must apply the simple random sampling within the different strata and more important is that the strata must be non-overlapping. Having overlapping means that some units will have higher chance to be chosen as subject. For example, to choose students introducing themselves at BA1, the teacher would first organize the class into groups like Asian, European, American and so on. After dividing students into groups, the teacher chooses randomly students from each group. By doing that, the teacher certainly does not miss any continent, which could happen when the teacher just used the simple random sampling. There are two types of stratified random sampling: Proportionate and disproportionate stratified random sampling. In proportionate stratified random sampling, each stratum has the same sampling fraction. Take dividing groups to introduce themselves as an example, the teacher chooses a sampling fraction of a half, this means a half of students each group are selected to introduce themselves. In disproportionate stratified random sampling, there are different...