The proportion of fat calories in the diets of people who read the nutrition labels on food products is significantly...

BenMingov on March 18, 2020

Hi Kenken, thanks for the question.

I am not sure that we can say that this is a correlation just because it doesn't come from a study or research. I think more accurately, and perhaps more easily understandable, is that in order to infer causation, you must truly prove the causation. Simply showing that two things/phenomena/events/whatever are in some way related, either positively or negatively does not mean that one causes the other.

In this situation, the proportion of fat in people's diets, comparing two different groups. Comparing those who read nutrition labels for calories and those who do not. The argument states that those who read the calorie content eat less fat. They conclude this is because of the reading.

While this may sound reasonable and intuitive, for all we know it is possible that because they eat less fat, they are now inclined to read the labels to continue this habit. We don't definitely know beyond a doubt that reading the label caused less fat consumption. It could be reversed. Additionally, it is possible that something external to both of these things causes reading the label and eating less fat, perhaps such as living in a warm climate, where you always have to be ready for a trip to the beach. The point is, who knows why this relationship exists? From this argument alone, we cannot conclude any causation. And also, don't forget, it is always possible for the two correlated items to be a coincidence, and there may be no causation.

So from now on, try thinking about causation skeptically. Unless it is proved, you can't infer it.

I hope this helps. Please let me know if you have any other questions.