In this presentation, we will discuss the concept of causality and will discuss how to examine a case from a legal perspective while looking for causal claims or relationships. The first area that we need to understand is the difference between causation and correlation. A correlation is a simple relationship between two items, events, or variables that occur together. This can be
based on a statistical analysis or on observation, but there is no implication that one caused the other.
In order to establish causality, you have to prove that one event produced another event as an effect. This is typically very difficult to prove and there are a number of different tests that need to be met to establish a causal relationship. The main one is that one must occur prior to the other – in other words, the cause must precede the effect. Second, the events have to be related, or covary.
Finally, other plausible alternative explanations must be ruled out, making
causality very difficult to prove. For example, when a pharmaceutical company wishes to establish a cause and effect relationship for a new drug (i.e., the effect is the treatment of a specific symptom or condition), it often takes years and dozens of studies to eliminate all other plausible explanations for the effect (i.e., the successful treatment of the condition). Asking the question “Why?” often leads us to the cause and effect relationship.
When we are examining causal claims, we need to keep in mind the conditions necessary to establish a causal relationship. Many times, an author will interpret a set of events and indicate that a causal relationship exists. In this case, we need to be able to rule out any other rival causes – in other words, we have to rule out any other potential interpretation or cause of the observed event. If we cannot rule out the potential rival causes, we will never know with certainty whether the author is correct in claiming the item is the true cause or if the rival event
(or even some other event) is the true cause of the observed effect.
There are three different types of rival causes: those related to differences between groups, those that have correlations between characteristics; and the post hoc ergo propter hoc fallacy.
First, any time an author claims that a cause stems from a specific difference between two or more groups, we have to ask ourselves, “Are there other differences in the groups that may be relevant?” If we think there may be another factor, we have a potential rival cause, thus reducing the strength of the author’s claim.
Second, there is simply a correlation between two items frequently, but the author claims there is a causal relationship between the items. Here, we should ask if the link works in reverse. In other words, let’s say the author claims that a company with a higher marketing budget causes higher performance, and a lower budget causes lower performance. While there may be a distinct  correlation between these two items, let’s ask ourselves the reverse question. In other words, “Can strong performance cause a higher marketing budget, and low performance.”