Causation vs Correlation

[By Zeal Akaraiwe]

I’ll give a quick example to illustrate the distinction between correlation and causation and, more importantly, the error in judgement that results from applying it incorrectly. When living in most parts of Lagos, it’s very easy to assume or conclude that “leaving the doors and windows open causes malaria,” and we were also told something similar as children: “If you hug a boy, you’ll get pregnant.” 

One way to think about correlation and causation is by acknowledging a crucial distinction between the two, which is: Causation is PROVEN, whereas correlation is OBSERVED.

The distinction is crucial for society as a whole since problems are resolved by addressing issues at the causality level rather than the correlation level. Therefore, it is simple for a highly opinionated and under-educated society to assume that mere correlation is a cause and therefore, expend energy on treating the symptoms of a correlation without ever addressing the cause. 

Here are two examples of what can happen when correlations are incorrect or when the correlation is addressed rather than the underlying reason. 

In an effort to relieve currency pressure, the CBN recently believed that felling trees in Abuja would increase the value of the naira. It goes without saying that after felling the trees, we were where we were. 

Later on, they asserted that anytime AbokiFX released or published currency rates, it led to a decline in the value of the currency, portraying the company as the adversary responsible for the depreciation of the naira. 

AbokiFX then stopped posting rates, but the decline in the currency persisted, just like when trees were taken down. 

It becomes incredibly clear why many of our societal concerns go unaddressed, and we start to believe that “super strong unseen forces” are to blame for the status of the nation. I have serious doubts about that and in my opinion, Hanlon’s razor is a good enough explanation. The source of all our evil is the persistent refusal of those in positions of power to address the pain’s primary causes. Alternatively stated, “vast ineptitude of wicked principalities in high places.” 

For instance, everyone is aware that “Lagos floods when it rains.” Since everyone generally notices this, there is no doubt that rain and flooding are closely related. However, can we draw the conclusion that rain actually causes floods given the strong correlation between the two? 

If the response to that question is “yes,” it implies that floods must occur whenever and wherever it rains, but I hope that we know that’s not the case. In addition, if we accept that the answer is “yes,” and since rain is something we cannot control, we subconsciously force ourselves to live with the outcome.

Floods are actually caused by bad or inadequate drainage systems combined with rain. Our drainage system mostly consists of roadside gutters, which are actually just storage areas for water rather than a drainage system. Therefore, the ongoing “cleaning” of these gutters will take a lot of time and resources, yet produce no results.

This is what is called contributory causality which is different from absolute causality. 

I’ll continue with the basic malaria example given in the beginning of this article to illustrate another type of causality: 

Living in a wet environment will induce you to OBSERVE that “open doors and windows promotes malaria,” and if you test it, you might even think it’s been proven. However, in other regions, if you reside on the 10th floor, leaving doors and windows open will result in better sleep and absolutely no malaria.

Malaria enters through open doors and windows. Therefore, there is a clear link between malaria and open doors (in regions where the female anopheles mosquito is present), but if the cause of the link cannot be determined, your treatment for malaria will end with “shut doors and windows.” There will be virtually little success from this. 

You must first address mosquito breeding as a fundamental malaria prevention measure before addressing mosquito access to humans. We’re not really that thorough or long-term oriented, though. Therefore, it appears that our problems are a never-ending cycle. 

This time, this is what is called conditional causality which is different from absolute causality.

These examples may help in explaining why we have numerous cases where people invest a lot of time, effort, and resources into a problem but see little or no progress. You may also have heard people say things like, “I’ve done my best” or “I’ve done all I can”,  but in reality, all they’ve done is waste money, time, and lives by addressing observed correlation when causation has not been determined.

The persistent cycle of “trial and error” creates a state of mind psychologists refer to as “learned helplessness,” which is defined as: “A state that occurs after a person or society has repeatedly experienced a stressful situation and then comes to believe that they are unable to control or change the situation, so they do not try — even when opportunities for change become available.”

After frequently going through difficult situations over prolonged periods, a person develops a sense of learned helplessness such that even when there are opportunities for change, individuals choose not to act because they begin to feel that they cannot influence or alter the situation. 

In conclusion, if a society suffers from incompetence for a long enough period of time, you will come to believe that there is no problem that can be solved, and if you ever had the chance to do so, you won’t even try. 

It then becomes understandable why, for many, if not most of our societal ills, the combination of inept leadership, being an overly religious society, our lack of ability to solve basic problems will lead us to accept that we are helpless and therefore, “ONLY GOD” can save us. We inadvertently have been forced to delegate solutions away from ourselves into the unknown. 

What will be the result when an unknown problem is taken to an unknown realm to receive unknown solutions?