In order to answer a particular research question, the researcher needs to investigate a particular area or group, to which the conclusions from the research will apply. The former may comprise a geographical location such as a city, an industry (for example the clothing industry), an organization/group of organizations, gender etc. This group is termed the research population.
Unless the research population is very small, we need to study a subset of it, which needs to be general enough to be applicable to the whole. This is known as a sample, and the selection of components of the sample that will give a representative view of the whole is known as sampling technique. It is from this sample that you will collect your data.
In order to draw up a sample, you need first to identify the total number of people in the research population. This information may be available in a telephone directory, a list of company members, or a list of companies in the area. It is known as a sampling frame.
Sampling may be done either a probability or a non-probability basis. This is an important research design decision, and one which will depend on such factors as whether the theory behind the research is positivist or idealist, whether qualitative or quantitative methods are used etc. Note that the two methods are not mutually exclusive, and may be used for different purposes at different points in the research.
PROBABILITY SAMPLING METHODS-
Probability sampling methods are said to be mathematical ways to select the group of people that will be interviewed. They are known as probability sampling methods because each person in the group from which the sample will be selected has a “known” chance (probability) of being selected. There are three probability sampling methods: Simple random samples, systematic samples, cluster samples and stratified samples.
Simple Random Sampling:
Simple random sampling involves making a list of the people from which the sample of respondents (people that will be interviewed) is selected. It ensures that every member of the population has an equal chance of selection. For example a group of 100 are listed and a group of 20 may be selected from this list at random. The selection may be done by computer or by hand.
The two basic procedures are:
1. the lottery method, e.g. picking numbers out of a hat or bag
2. the use of a table of random numbers.
Uses: Simply to design and interpret; can calculate estimate of the population and the sampling error.
Limitations: Need a complete and accurate population listing; may not be practical if the sample requires lots of small visits all over the country
This method is often used instead of random sampling. Just like simple random sampling, systematic sampling involves drawing up a list of potential respondents. The next stage is to decide what system will be used to select the respondents from the list. For example every 5th person on the list will be selected as a respondent so that would be the 5th, 10th, 15th, 20th person and so on. Similarly if the interviewers implement a 10th person system, they would choose the 10th, 20th, 30th, 40th, 50th person and so on. This method is simple but remains hard to follow-up and to check.
Uses: Easier to extract the sample than via simple random; ensures sample is spread across the population
Limitations: Can be costly and time-consuming if the sample is not conveniently located
Units in the population can often be found in certain geographic groups or “clusters” (e.g. primary school children in Derbyshire. A random sample of clusters is taken, and then all units within the cluster are examined).
Uses: Quick & easy; does not require complete population information; good for face-to-face surveys
Limitations: Expensive if the clusters are large; greater risk of sampling error
This technique is similar to the simple random sampling method described above except the researcher divides the population into subgroups first (for example by certain demographic groups of interest), then randomly samples members from each sub group.
Uses: A more precise method of sampling means you don’t need as high sample size. You are less likely to have an unrepresentative sample.
Limitations: Can be time-consuming and take more administrative effort than simple random sampling.
NON PROBABILITY SAMPLING METHODS-
These are techniques in which sample is not randomly selected. That is, not every member of the population has an equal opportunity of being chosen to participate in the survey.
Where the researcher questions anyone who is available This method is quick and cheap. However we do not know how representative (of the population) the sample is and how reliable the result. It is down to the researcher to ensure that they pick a people with a large variety of characteristics. An example of this is surveying students in one or two classrooms that the researcher has been allowed access to. Being present at a particular time e.g. at lunch in the canteen. This is an easy way of getting a sample, but may not be strictly accurate, because the factor you have chosen is based on your convenience rather than on a true understanding of the characteristics of the sample.
Uses: Respondents are readily available and a large number of surveys can be completed in a relatively short period of time.
Limitations: Cannot extrapolate from sample to infer about the population; prone to volunteer bias.
Using this method the sample respondents are made up of potential purchasers of your product or the market that you would like to research. The sample will contain people who meet certain criteria e.g. age, social group, gender. Splitting the sample into age is a popular way to apply quota sampling, the researcher will be asked to interview a set number (quota) of people from different age groups e.g. 16-25, 25-40, 40-55, 55 and above. Quota sampling ensures that the sample contains people satisfying all of the characteristics in the market being researched.
Uses: Quick & easy way of obtaining a sample
Limitations: Not random, so still some risk of bias; need to understand the population to be able to identify the basis of stratification
Quota v random sampling
The advantages and disadvantages of quota versus probability samples has been a subject of controversy for many years. Some practitioners hold the quota sample method to be as unreliable and prone to bias as to be almost worthless. Others think that although it is clearly less sound theoretically than probability sampling, it can be used safely in certain circumstances. Still others believe that with adequate safeguards quota sampling can be made highly reliable and that the extra cost of probability sampling is not worthwhile.
Generally, statisticians criticise the method for its theoretical weakness while market researchers defend it for its cheapness and administrative convenience.
A deliberate choice of a sample – the opposite of random In judgement sampling, the researcher or some other “expert” uses his/her judgement in selecting the units from the population for study based on the population’s parameters.
This type of sampling technique might be the most appropriate if the population to be studied is difficult to locate or if some members are thought to be better (more knowledgeable, more willing, etc.) than others to interview. Statisticians often use this method in exploratory studies like pre-testing of questionnaires and focus groups. They also prefer to use this method in laboratory settings where the choice of experimental subjects (i.e., animal, human, vegetable) reflects the investigator’s pre-existing beliefs about the population.
This determination is often made on the advice and with the assistance of the client. For instance, if you wanted to interview incentive travel organizers within a specific industry to determine their needs or destination preferences, you might find that not only are there relatively few, they are also extremely busy and may well be reluctant to take time to talk to you. Relying on the judgement of some knowledgeable experts may be far more productive in identifying potential interviewees than trying to develop a list of the population in order to randomly select a small number.
Uses: Good for providing illustrative examples or case studies
Limitations: Very prone to bias; samples often small; cannot extrapolate from sample
In snowball sampling, you begin by identifying someone who meets the criteria for inclusion in your study. You then ask them to recommend others who they may know who also meet the criteria. If you have a small or rare population, or do not know who your population is e.g. hard drug users in your city, this may be the only way to achieve a sample.
For instance, if you are studying the homeless, you are not likely to be able to find good lists of homeless people within a specific geographical area. However, if you go to that area and identify one or two, you may find that they know very well who the other homeless people in their vicinity are and how you can find them.
To create a snowball sample, there are two steps: (a) trying to identify one or more units in the desired population; and (b)using these units to find further units and so on until the sample size is met.
Uses: Useful when you are trying to reach populations that are inaccessible or hard to find
Limitations: Many a time this method would hardly lead to representative samples.
Other Members: Amrit Jain, Ankit Saxena, Gugan N, Jyoti Kanwatia, Nitin Sonkar, Sonam Supriya, Yogesh Sham Gupta