Case and Class Probability

Let’s imagine there is a disease or virus that is spreading in which 98% of those who have contracted it have survived. Although it may be comforting to hear, suggesting that if one does contract the bug they as an individual have a 98% chance of surviving would be a misnomer. It would also be fallacious to suggest that if 98% of others in a group contracted the illness that there is no longer a chance for the other 2% to contract it. While the latter example is a common gambler’s fallacy, the former example is an important distinction to be made between ‘case’ probability and ‘class’ probability.

Case probability looks to a particular event through a lens of understanding what some of the causes were that contributed to its teleological outcomes (110). In order to remember this first distinction think about case studies in an MBA program examining the outcome of a particular marketing strategy, for example. Where the case is unique in and of itself, it is not deterministic of all other cases no matter the similarities. Past examples, or cases, do not determine future outcomes, especially when it comes to human action or involvement in a constantly shifting world.

Class probability, also known as frequency probability, according to Dr. Ludwig von Mises, means “we know or assume to know, with regard to the problem concerned, everything about the behavior of a whole class of events or phenomena; but about the actual singular events or phenomena we know nothing but that they are elements of this class,” (107).

For an epidemiologist, or even an economist, the importance of this distinction between case and class is that once these terms of ‘probability’ are accepted it is evident that “understanding is the only appropriate method of dealing with the uncertainty of future conditions” necessarily (118). This means that, although ‘probability’ exists, those wielding probability cannot be omniscient in knowing outcomes, instead probability helps to understand. ‘Understanding,’ by its very nature, “is always based on incomplete knowledge,” (112). Simultaneously, the understanding required is beyond the numbers or the symbolized results of a study, it must include the understanding of the circumstances; even one’s philosophy influences their scientific study and its results (643). ‘Probability’ denotes that there is a possibility of variance, yet quite often probability in conjunction with statistics, is erroneously referred to as though deterministic and universal. ‘Case’ and ‘class’ probability help to better understand ‘probability’ without imposing hard determinism and sweeping declarations.

So, if there is a disease or virus that is spreading in which 98% of those who have contracted it have survived, this does not mean that each person who contracts it has a 98% chance of survival. It also does not mean that the other 2%-factor is not influenced by a philosophy or perspective. Each metric has its philosophical grounding and definition, this does not mean that everyone’s is the same perspective or method, thus it is neither a universal nor a deterministic fact. The ‘factual’ conclusions, or results, may be completely different in two studies who use different philosophies and meanings behind the words or conditions selected. The case-class distinction helps to recognize these limitations of probability through understanding while not resorting to begging the question.