Correlation Vs Causation
Two things move together. The monsoon rolls into Lahore in July and August, and dengue cases at Mayo Hospital rise in the same months. Did the monsoon cause the dengue, or are they just dancing in the same room. This is…
There is more than one way to understand this. If you have only been taught one, you have been taught less than you deserve.
There is more than one way to understand this. If you have only been taught one, you have been taught less than you deserve.
Two things move together. The monsoon rolls into Lahore in July and August, and dengue cases at Mayo Hospital rise in the same months. Did the monsoon cause the dengue, or are they just dancing in the same room. This is the difference between correlation and causation, and it is one of the most useful distinctions a thinking person learns.
Correlation just means two things move together in the data. Causation means one of them actually pushes the other. Dengue rises in monsoon months because the mosquito that carries dengue, Aedes aegypti, breeds in standing water. The rain creates the standing water in tyre dumps, plant pots, and unused buckets. The chain is rain to standing water to mosquitoes to dengue. There is real causation, and there is a mechanism, which is what tells us this is not a coincidence.
Now a trickier example. Students who go to private tuition in Defence and Gulberg score higher in matric than students who don't. Did the tuition cause the higher score. Maybe. But a quieter explanation also fits: families who can afford 15,000 rupee a month tuition in DHA also have parents who can read English, can sit with the child every evening, can buy stationery without thinking about it, and can keep the child away from family chores. Tuition and high scores both ride on a third thing: family income. The third thing is called a confounder.
A funny example to lock it in. Kiryana shop sales of Pepsi and chips in Lahore neighbourhoods spike during a Pakistan vs India cricket match. Did Pepsi cause the cricket match? Did the cricket match cause Pepsi sales? In this case, the cricket match is the cause and the sales are the effect; people stock up to watch the match together. The mechanism is obvious. Now flip it. If you noticed that fan sales in Karachi rise in the same week that ice cream sales rise, would you say fans cause ice cream cravings? Probably not. The third thing is summer heat. Both are effects of one cause.
The real world rarely runs on one cause. Dengue is rain plus standing water plus mosquito plus unprotected skin plus a virus already in the city. Matric scores are tuition plus parent literacy plus sleep plus nutrition plus a school that actually teaches. A thinking Pakistani learns to look for the bunch of causes, not the one cause that fits in a tweet.
Estimated time: 14 min