Bunuel wrote:
Recent studies have demonstrated that smokers are more likely than nonsmokers to develop heart disease. Other studies have established that smokers are more likely than others to drink caffeinated beverages. Therefore, even though drinking caffeinated beverages is not thought to be a cause of heart disease, there is a positive correlation between drinking caffeinated beverages and the development of heart disease.
The argument’s reasoning is most vulnerable to criticism on the grounds that the argument fails to take into account the possibility that
(A) smokers who drink caffeinated beverages are less likely to develop heart disease than are smokers who do not drink caffeinated beverages
(B) something else, such as dietary fat intake, may be a more important factor in the development of heart disease than are the factors cited in the argument
(C) drinking caffeinated beverages is more strongly correlated with the development of heart disease than is smoking
(D) it is only among people who have a hereditary predisposition to heart disease that caffeine consumption is positively correlated with the development of heart disease
(E) there is a common cause of both the development of heart disease and behaviors such as drinking caffeinated beverages and smoking
EXPLANATION FROM Fox LSAT
No, no… you haven’t proven there’s a correlation between anything and anything else. I’m not buying it.
This argument basically proceeds like this:
- Premise 1: “A (smoking) and B (heart disease) are correlated.” I have no problem with this. It’s a premise of the argument, and we should accept it as fact.
- Premise 2: “A (smoking) and C (drinking coffee) are correlated.” I have no problem with this either. It’s a premise of the argument, and we should accept it as fact.
- Conclusion: “B (heart disease) and C (drinking coffee) are correlated.” I have a huge problem with this. It’s not a premise, so we should not accept it as fact. It’s the conclusion of the argument, which is usually where the bullshit comes in. Here, the conclusion is suggested by the facts, but definitely not necessarily proven by the facts. So we object. It might be true, but the given facts certainly don’t prove it.
An example might clarify why B and C might not necessarily be correlated. What if the argument had said this:
- Premise 1: “A (living in San Francisco) and B (seeing a buck naked dude walking down the street in broad daylight) are correlated.” (This is true.)
- Premise 2: “A (living in San Francisco) and C (being wealthier than average) are correlated.” (This is also true.)
- Conclusion: “Therefore B (seeing a buck naked dude walking down the street in broad daylight) and C (being wealthier than average) are correlated.” (This is obviously nonsensical.)
How’s that? Make more sense? General principle: Just because A and B are correlated, and A and C are also correlated, does
not mean that B and C are correlated.
The question asks us criticize the argument by finding a possibility “that the argument fails to take into account.” I think the answer might be something like “B and C are negatively correlated,” or “B reduces C,” or “C reduces B.”
A) Yep. This is “C reduces B.” If it’s true that smokers who drink caffeine are
less likely to get heart disease than are smokers who do not drink caffeine, then that’s a pretty good attack on the idea that caffeine and heart disease are correlated. I like this answer a lot.
B) Nah. Who gives a **** if there is a “more important factor”? A more important factor wouldn’t do anything to weaken the idea that B and C are correlated. B and C can still be correlated, even if X and B are even more correlated.
C) This would strengthen the argument. We’re looking for a weakener.
D) Even if this is true, it doesn’t weaken the argument. Actually, it strengthens the argument by saying that, for at least one segment of the population, there
is a correlation between caffeine and heart disease. We wanted to weaken that correlation, not strengthen it.
E) A common cause between B and C would
strengthen the idea that B and C are correlated.
Our answer is A.
But the option hinges on the likelihood, meaning the degree may vary from “more to less” but it doesnt change the fact that it leads to it, no matter how small the effect is. Isnt that strengthening the fact that caffeine causing heart disease? Say in smokers it may be only 50% or say 90% of the times true but in no -smokers it may be 95%. But that doesnt mean it doesnt.