Formal sample size issues are essential only in academic (enumerative) statistics, while in analytic statistics, the answer to “How much data?” is “Enough to characterize the underlying situation.” This is because, in enumerative statistics, the elements of the population are well-defined and constant in the study.
Furthermore, according to Deming, the population can sometimes be an already known finite population which the researcher would want to conclude. In other words, the purpose of such statistics is identifiable. Also, Nash et al. illustrate that the sample points from which data is to be collected can either be obtained from the definition of the study population or available to the researcher.
Additionally, analytical statistics conclude that the process is non-existent at the time of the study. In short, the objective of analytic statistics is to improve future operations. Moreover, the study lacks a well-defined sampling frame in analytical statistics, while the impact is short-term and localized.
Academic/research statistics are fundamentally different from analytic statistics because their interest is based on the group or material from which a sample is taken. As a result, the study is descriptive. For instance, it can be used to determine how many people belong to a given category.
Its core intention is not to invest in the process of determining the outcome. Additionally, the decision of the actual study is made based on the material. However, in analytical statistics, the focus is on the process that produces the outcome. Its main aim is to improve the practice being investigated in the future.
As a result, it helps in predicting the future through the process. As opposed to enumerative statistics, decisions in analytical studies are made to alter or maintain the process. Additionally, the data obtained in analytical is used to compare processes to make decisions by outweighing the benefits and weaknesses of every process being investigated.