The Little Blue Reasoning Book - Brandon Royal [31]
As depicted in Exhibit 4.1, cause-and-effect relationships arise under six potential categories. These include the following:
Exhibit 4.1 – Coincidence, Correlation, and Causation
The first question we ask when a cause-and-effect assumption is on the horizon is whether any relationship exists between two items. There may not be any plausible relationship. For example, “The street light turned red just before the cat fell out of the tree; therefore, the red light caused the cat to fall out of the tree” has no plausible causal relationship (mere coincidence). Next, assuming a relationship exists, we ask whether the two events are causally related or merely correlated. If a correlation exists, we seek to determine whether that correlation is low or high. If causally related, we seek to determine whether the two events are legitimately correlated or whether alternative or reverse causation is at work.
Here are further explanations of the categories highlighted in the previous chart.
I. Mere Coincidence
“Every time I sit in my favorite seat during a playoff game, our team wins.” It is unlikely that your “lucky” seat is causing your sports team to win. And it is equally unlikely that a regular or “bad” seat will cause your team to lose.
II. Low Correlation
An example of low correlation might be the opening of new health clubs in your city and the general level of fitness among citizens in your city. Obviously, the opening of health clubs with facilities that include weight-lifting classes, aerobic classes, and exercise machines will have some effect on the fitness level of people in general. But, practically, it will not have a great deal of impact. The direct impact of a small number of health club members on a city’s larger population is limited. Even if there is a general trend toward more fitness in your city, it may be because people walk, ride bikes, and take hikes more often. Individuals may participate in these activities and not be associated with health clubs.
III. High Correlation (but not causation)
Certain factors or characteristics are strongly correlated — for example, being tall and being a National Basketball Association player. Not every player in the NBA is tall, but the vast majority of players are. We can safely say that there is a strong correlation between being tall and being an NBA player. A classic example in business is a company’s sales and advertising costs. The more a company spends on advertising, the greater its sales. (The correlation between advertising and sales is approximately +0.8.) Other examples might include hot weather and ice cream sales, or rainy weather and umbrella sales. Strongly correlated events may be talked about as if they are causally related. It is important to be able to draw the line between high correlation and actual causation.
IV. Legitimate Causation
The law of gravity indicates a causal relationship. I throw an apple up into the air and it comes back down. Other events are so highly correlated that for all practical purposes they are assumed to be causally related — for example, the amount of coffee consumed and the amount of coffee beans consumed or the number of babies born and the number of baby diapers used. However, it would not be accurate to say the number of coffee beans grown or the number of baby diapers manufactured.
V. Alternative Explanation
Alternative explanation technically can be called alternative causal explanation. Here we agree on a single conclusion (the effect) but differ as to which is the correct cause. We must always be on guard for the existence of another cause whenever it looks as though two events are otherwise causally related. A business may have increased its advertising budget and seen an increase in sales. It is easy to view these two events as causally