View Single Post
Old 09-24-2017, 01:36 AM   #145
loopytheone
Administrator
 
loopytheone's Avatar
 
Join Date: Dec 2012
Location: Robinhoodland
Posts: 3,350
loopytheone has ascended what used to be the highest level.loopytheone has ascended what used to be the highest level.loopytheone has ascended what used to be the highest level.loopytheone has ascended what used to be the highest level.loopytheone has ascended what used to be the highest level.loopytheone has ascended what used to be the highest level.loopytheone has ascended what used to be the highest level.loopytheone has ascended what used to be the highest level.loopytheone has ascended what used to be the highest level.loopytheone has ascended what used to be the highest level.loopytheone has ascended what used to be the highest level.
Default

Quote:
Originally Posted by TwoSwords View Post
I love that comparison. In this example, the causation is, at best, indirect, and in the opposite direction, and the same can be true in the case of most correlations. This is why the claim that obesity causes poor health is on such shaky ground. It could be like your fire truck example, and obesity could be just another symptom of poor health, or it could be only one of many consequences arising from some other factor. For instance, it could well be that obesity is merely a common symptom of over-consumption of refined sugar and lack of exercise. Excessive stress and lack of sleep, of course, worsen any condition too. We know these things are bad for our health, and blaming obesity for conditions that also have these factors in common is like saying that people with hair loss are more likely to catch radiation poisoning.
I don't know if you have taken any statistics classes or anything, but were you ever given the example about ice cream and murders? It's a classic example used to explain to students why a strong correlation is not the same thing as cause and effect. The example is that if you were to plot a graph of ice cream consumption on one axis and murder rates on the other, you'd see a really strong correlation. Does that mean that ice cream makes people murderous, or that all murderers love ice cream? Of course not, the reason they correlate is because of the weather; people eat more ice cream in summer and more murders happen in summer. That's always my go to example when trying to explain the difference between correlation and cause and effect to people.

But yeah, you have exactly the same thoughts/wariness when it comes to these correlations around obesity as I do.
loopytheone is offline   Reply With Quote