Pause, before you pin the cause

Our brains have evolved over time to detect threats to our existence by looking for patterns in our environment and extrapolating from those patterns. The reptilian part of our brain still looks for fight or flight responses. Hence, for animals in the wild, the sight of bushes moving could imply that there is a stiff breeze or else a predator waits for its prey. The safer choice for them is to fly to safety even if it is just the wind, since, staying put can lead to their death. Thus, it is safer on the whole to assume causation as opposed to correlation between events.

As humans, while our bigger brains have adapted over time to fine tune these flight or fight responses, we continue to make the same mistakes with regard to attributing causation in the face of simple correlations. What do I mean by that? Think of the number of times when we are in an argument, say regarding the utter silliness in believing that a black cat crossing the road is an ill portent, and someone gives a personal anecdote about how in their case, a black cat had indeed led to their miserable failure in an exam. This is a logical fallacy known as ‘Post hoc, ergo propter hoc’, meaning “after this, therefore because of this”. All of us have done this at some time or the other. However, what gets my goat is the number of times I see similar arguments uttered by journalists and politicians.

A couple of recent examples come to mind. The first story was about fifty odd people dying “because of demonetization”. Just stop to think about it. In a country of 1.2 billion people, you could link fifty odd deaths to anything, even to India’s win over England in a cricket match. However, that wouldn’t make much sense and so it is more expeditious for politicians to use such numbers to bolster their own arguments. The second example was a headline in a leading newspaper this morning that said “26 die, unable to bear grief of separation”, attributing the cause of these deaths to the demise of the Tamil Nadu CM Jayalalithaa. This is again a case of using a simple correlation and transforming that into a cause and effect.

The problem with conflating correlation with causation is that anyone can pull up any number of spurious cause-and-effects using correlations. There is a classic example of an almost perfect correlation between the US spending on science, space and technology and suicides by hanging, strangulation and suffocation. Or that between per capita cheese consumption and the number of people who died by becoming tangled in their bed sheets. The website www.tylervigen.com has a whole list of such spurious correlations.

So next time we see such news items, it would be best to pause and think as to whether these actually are causal relationships or mere correlations.

The writer is founder and CEO of HR analytics start-up nFactorial Analytical Sciences