Data collection is an essential part of the decision making process. In order to avoid subjective variables in the process, it is essential to gather objective information on which the decision will be based on. Sampling is one of the methods that enable to collect data. The aim of this method is to yield some knowledge about the population, by the decision made. For instance, when launching a new advertising campaign, a company needs to gather information on its customers' buying habits, consuming motivators and so on. Sampling is consequently a way to study a smaller, representative sample of the population to gather information. By definition, sampling is "the process of selecting units (e.g., people, organizations) from a population of interest so that by studying the sample we may fairly generalize our results back to the population from which they were chosen."
In this report, we will focus on the key issues regarding "sampling" in the managers' decision making process. After analyzing the different sample selection techniques, we will study the way business managers design strong samples statistically that can be used for multiple purposes. Finally, we will evaluate the way business managers adjust the sample designs so that the samples remain representative as organizational characteristics change over time.
The purpose of this report is to explain how sampling can be used as a reliable tool in the decision making process, even though this process can be endangered by managers' subjectivity and the environmental changes.
[...] - High subjectivity: You choose on what variable you base the quota on Snowball sampling (or referral sampling) Snowball sa mpling c onsists in bu ilding t he p ortrait of t he t ypical individual t hat sh all co mpose your sample. T hen, you hav e to i dentify one per son w ho meets those cr iteria an d a sk him t o recommend other persons, also meeting those criteria, and so on. Even if it is sometimes the best alternative, this method often fails to objectively represent the overall population. [...]
[...] In order to constantly adapt (the sample) to ensure that it remains representative, business managers have to be on constant watch: they must be aware of new regulations, policies, social trends, economic situations, etc. Internet is a perfect tool to follow those evolutions, but business managers also have to develop networking to share information with fellow workers and to be able to anticipate evolutions. In order to record those environmental changes, business managers also have to regularly update numerous databases, so that information is recorded and can be found easily whenever needed Identifying variables Each sample is designed according to the specific variables: a target population is selected and a sample is designed through a probabilistic or non-probabilistic process. [...]
[...] The more general the sample is, the more it can be used in business research. For instance, the brand Coca Cola may carry out a market survey among its customers. To do so, the company has to select a sample of customers in its core target. If this sample is statistically strong enough, the sample could be used several times for different researches. This is the reason why designing a strong sample can enable the company to spare a lot of money. [...]
[...] One important thing would be to avoid having periodicity between qualifications in the list (doctors, nurses) so that selected units would not be all doctors Cluster random sampling Cluster random sampling was invented to deal with geographical issues in sampling. For instance, a marketing c ompany carrying out a s urvey might b e r eluctant to interview p eople from everywhere in France because the cost would be very high. With the help of this method, samples are “clustered” so that the process is cheaper. [...]
[...] The purpose of this report is to explain how sampling can be used as a reliable tool in the decision making process, even though this process can be endangered by managers' subjectivity and the environmental changes. There are two main sample selection techniques: Probability sampling and nonprobability sampling Probability sampling Probability sampling refers to a m ethod considering that each unit of the population has equal probabilities of being selected as the sample. This method therefore uses random s election ( picking a na me ou t o f a ha t). [...]
Online readingwith our online reader
Content validatedby our reading committee