agdimple333 wrote:
Citing the frequency with which gum disease and heart disease occur in the same patients, many dentists believe that periodontal disease is a cause of a variety of cardiovascular problems, including Coronary Artery Disease.
Which of the following, if true, would most weaken the claim that periodontal disease is a cause of Coronary Artery disease?
A. Bacteria present in infected gums can become mobile and enter the bloodstream, causing arterial plaque to accumulate.
B. People who brush and floss their teeth regularly are also more likely to exercise and eat a healthy diet.
C. Infected gums are more prone to bleeding, which allows bacteria to escape the mouth and irritate arteries.
D. People who experience loss of teeth due to periodontal disease usually cut back on many foods that are harder to chew, such as lean meats and vegetables, and increase their consumption of processed foods like pudding and ice cream.
E. Patients with no history of heart disease are much less likely to have periodontal disease than patients who have had a cardiac transplant.
Another strange question... Is it just me or do you guys also think these questions are not true representation of actual GMAT style questions..
fameatop wrote:
I am not able to understand why option B is correct & E is incorrect. Can you kindly throw some light on the same. Waiting eagerly for your detailed explanation.
So far as I can tell, this is a Knewton question, and the OA is
(B). First, as to
agdimple333's point, while this prompt is perhaps a little on the short side, I would say the logic of the question very much captures the kind of logic you will see on GMAT CR questions. This is, in essence, a very good question --- not least because it has a very clear and well-defined OA, and yet, many folks on this page have fallen for one of the trap answers, most notably,
(E).
The big underlying idea of this question is ----
correlation does not imply causality. This blog
is primarily about regression and correlation, but it does touch on this issue.
Let's say
P = periodontal disease
Q = cardiovascular problems
The dentists' argument is, essentially, P & Q are correlated, so P cause Q. The dentists' argument is abysmally bad, a classic flawed argument pattern.
To
strengthen such an argument, we would have to demonstrate there was some mechanism of connection ---- e.g. bacteria or viruses in the mouth that become blood-borne and infect the heart, something like that.
One of the best ways to weaken a correlation/causality argument is to show that both terms arise from something else. This is in essence what
(B) does.
(B) says: there are a category of people --- call them "health all over" people --- and these folks take care of themselves from head to foot --- they brush and floss, which prevents periodontal disease, and separately, they eat healthy and exercise, so they don't get heart disease. That implies there would be other people, the "don't take care of self" people, who don't brush, don't floss, don't eat health, don't exercise, don't laugh, don't sing, don't do much of anything to take care of themselves. These latter people would be prime candidates to get periodontal disease (from not brushing & flossing) as well as heart disease (from poor diet and no exercise), but the periodontal disease and the heart disease do not have relationship of
causality with one another --- rather, they are both
products of an overall unhealthy lifestyle. This decisively weakens the argument, which is all about the leap from correlation to causality.
By contrast,
(E) simply provides more evidence for the correlation. We already know P & Q are correlated. That was the first sentence. That's the evidence in this argument. That's beyond doubt. The crux of the argument is this vast logical leap from correlation to causality. Choice
(E) simply gives more evidence that P & Q are correlated. In a strange way, it is a kind of strengthener, insofar as it reinforces evidence. It doesn't address the issue of causality and it doesn't clearly weaken the argument.
Another way to weaken the argument "P causes Q" would be to show that, actually, Q causes P --- something along those lines would be a good weakener, but
(E) doesn't clearly support this.
Does all this make sense?
Mike
Thanks Mike, this explanation was clear. But reading it gave me the impression that the question is not representative of what Official GMAT would look like. GMAT answers wouldn't require any sort of "over" inference, but answering B was necessary to infer something beyond, as you've pointed out:
In other words, to choose answer B is necessary to assume that healthy habits and exercise prevent someone from getting the heart disease (real world true, but the question really implied that?). This is what refrained me from picking B.