To succeed in any market, you need to be well prepared. Being prepared means having as much data on a prospective market as possible. As the old saying goes, knowledge is power.
Sampling is one of the primary ways to gather the important data you will need in order to be successful in a new market. The FITTskills Feasibility of International Trade course delves into the best methods for researchers to create a representative sample.
For researchers to make conclusions about the population from the sample, it must be representative.
This means the sample group should include the same major characteristics of the larger group, as well as be chosen in a way that will reduce the likelihood of selecting members who are not representative of the population as a whole.
To accurately create a representative sample, researchers must ask themselves a simple question.
What Sampling Methods Will Be Used?
After defining the sample characteristics, the researcher must choose members of the sample to participate in the study. To do this, researchers use either of the following:
A probability based method
This is a random selection method, in which respondents in the sample pool each have a pre-established chance of being selected for the survey.
A non-probability based method
This is a selection method in which respondents are chosen because of convenience, judgment, or through quota methods.
The probability sample provides data that can be statistically tested to identify the degree of its similarity to the larger population. Even so, there may be errors that have nothing to do with the sampling method. This possibility should be considered when evaluating the cost/value trade-off of using a probability versus a non-probability based sample.
Probability based methods of sample selection can take a great deal of time and expense that can be magnified when applied internationally.
To benefit from the precision of these methods, a researcher must be sure the sampling tool is complete and current, as well as be sure that the individuals who are randomly chosen are accessible. The expense of this approach is justified by the benefits gained.
The non-probability based sample is often selected because of cost, time, and value considerations, or because the entire population might not be available to the researcher. The ability to identify and access an unbiased random sample might be hampered by cultural factors or logistical problems. In such cases, the precision of the probability based sample is lost, making the high cost difficult to justify.
If the survey is being used to gather more general ideas or trends, a non-probability based sample can be adequate, if it is carefully chosen. The researcher should understand there is no method to statistically test whether it is representative of the entire population.
There are four generally accepted methods of selecting a non-probability based sample:
1. Convenience
This is the least reliable, the least expensive, and the easiest method. The researcher chooses whoever is close and easy to access. For international trade research, this method might be difficult to implement if the company has no previous contact in the market, because selecting people who happen to be passing by might not be acceptable. An example of the convenience method is taste-testing in a supermarket.
2. Judgment
This is a more reliable method than convenience because respondents are chosen based on some pre-defined characteristic. An example is the use of key clients to test changes to a product.
3. Quota
This involves a greater effort to define the sample, but offers higher reliability and a greater degree of similarity with the characteristics of the population. at large. The researcher specifies the characteristics of value to the research question and selects a group that has the same proportion of these characteristics as the larger population.
4. Sample size
As with the method of choosing the sample, the researcher can use either a qualitative or a statistical approach to selecting the sample. The choice depends on the degree of accuracy and the amount of acceptable risk.
There are several methods of gathering primary data. Focus groups are a very effective way of obtaining qualitative data, while surveys and questionnaires can deliver qualitative or quantitative data, depending on how the survey is designed. Developing questions for surveys, interviews, and focus groups requires skill and experience. Poorly worded or structured questions will produce inaccurate results.
The method chosen for a research project will depend on the type of data that is required and the research objectives that must be met.
An essential consideration for primary research is selecting a sample to survey. It is too impractical for companies to survey an entire population to gather research, but it is important to select a sample best matching the population at large in order to gain the most realistic results.
Are there more factors to consider when making sample group that we haven’t listed? Share them with us in the comments down below.
disqus comments