Q20 - A person with a type

[ohthatpatrick]

Question Type:
Inference (must be true)

Stimulus Breakdown:
Type B has higher risk of heart disease than type A. When put on a low cholesterol diet, type B's chol went down but lipid profile stayed same. Type A showed no benefit, and 40% of them actually became type B.

Answer Anticipation:
Inference questions want us to combine facts, usually via Conditional, Causal, or Quantitative language. Here we could combine the first sentence and the last sentence and say that a bunch of Type A people were MADE WORSE (causal) by going on a low cholesterol diet. It switched them to type B, which has a higher risk of heart disease. And we were told they received no benefit from the diet. So we could prephrase something like "type A people shouldn't go on a low fat diet" or, using safer language, "going on a low fat diet is a net loss for some type A people".

Correct Answer:
E

Answer Choice Analysis:
(A) Extreme: most. Can we say that over 51% of the volunteers had their heart disease risk lowered? Heavens no. We have no idea what the respective totals were for type A and type B volunteers. The only thing we were told that would lower heart disease risk would be switching to type A, and no one did that.

(B) We can't compare the cholesterol levels of these two groups. We know that B's cholesterol went down, but it may have started at a lower/higher/similar level compared to A's cholesterol.

(C) We can't say anything for sure about whether these volunteers changed other aspects of their lives.

(D) Extreme: solely. This ties into choice (C). We don't know whether the volunteers changed anything else in their lives duirng the experiment.

(E) Yup! This is what we predicted by putting facts together. The type A people who became type B had their heart disease risk increase.

Takeaway/Pattern: The correct answer has really safe language: "for at least some" (at least one).The first sentence gives a general relationship between Type B and A. Then we hear a description of an experiment. Our Inference brain should be attempting to apply what we know from the 1st sentence's generalization to what we hear about the experiment.

#officialexplanation

[mrudula_2005]

Clearly by POE, E is the best answer. But why "must [it] be true" ?!
Doesn't it illicitly presume causation from mere correlation? All that we know from sentence 1 of the stimulus is that "a person with a type B lipid profile is at a much greater risk of heart disease than a person with a type A lipid profile." - that is just a correlation, right?

So how can we say that (E) must be true based on the fact that 40% of the type A volunteers shifted to type B profiles?! To me, all that we can say is that For at least some of the volunteers in the experiment, they are now included in a group with a typically much greater risk of heart disease.

help? thanks!

[aileenann]

Hi there!

I think you actually stole the words out of my mouth. (E) must be true - that is it must be true that for at least some volunteers (namely those who went from the better type A profile to the riskier type B profile) are now at greater risk than they were.

I think what you are expressing is a discomfort with this probabilistic thinking. We can't point to a particular person who is now going to have a problem as a type B profile but who wouldn't have before the low fat diet. But remember, the answer choice (E) is only phrased probabilistically as well - they refer to an "increased" risk - not of a particular person at risk.

Does that make sense? Let me know

As for the wrong answers:

(A) is contradicted - the folks that switched to type B had their risk increase!

(B) is tempting, but just because type B folks are at higher risk, we don't know if they have more cholesterol - don't bring in outside knowledge (and by the way, isn't there good and bad cholesterol?)

(C) is tempting if you thought the stimulus was saying the results were inconclusive - but all we hear are the results.

(D) is too strong! Solely? How do we know there wasn't something else that affected the cholesterol levels?

[mrudula_2005]

Hi! Thanks for your response(s)! I still have the same concern about (E) though - maybe I misunderstood you, but in (E) the fact that they say "for at least some of the volunteers" does necessitate at least one particular person at increased risk, and even if not that wasn't really my concern...i just don't see how they are jumping from a correlation that ppl with type B lipid profile are at a much greater risk of heart disease than people with type A to a flat out definitive statement that at least some of the volunteers (the ppl who switched from type A to type B...or at least some of the people in that group) absolutely experienced an increase in the risk of heart disease due to their switching to type B profiles. I just don't see how that must follow.

thanks again for all your help!

[melnauni]

I also don't see how we can jump to the conclusion without thinking that profile B somehow influences the risk. Isn't it like saying people with blonde hair are more likely to prefer ice-cream than people with brown hair. Therefore, if you dye your hair blonde, you are more likely to prefer ice-cream?

[chike_eze]

Hi! Thanks for your response(s)! I still have the same concern about (E) though - maybe I misunderstood you, but in (E) the fact that they say "for at least some of the volunteers" does necessitate at least one particular person at increased risk, and even if not that wasn't really my concern...i just don't see how they are jumping from a correlation that ppl with type B lipid profile are at a much greater risk of heart disease than people with type A to a flat out definitive statement that at least some of the volunteers (the ppl who switched from type A to type B...or at least some of the people in that group) absolutely experienced an increase in the risk of heart disease due to their switching to type B profiles. I just don't see how that must follow.

thanks again for all your help!

Let me take a stab at this!

Type A higher risk of heart disease than Type B.
Both Type A and Type B in experiment e.g. 5A, 4B
Type B = lowered cholesterol, but still Type B
Type A = some Type A people switched to Type B
therefore,
Some Type A (in the study) became Type B, which means at least some people's risk for heart disease (in the study) increased.

I think, without over-analyzing the stimulus, (E) simply states that some people with Type A (e.g. 20% risk), became Type B (e.g. 60% risk). Therefore, the risk (probability) for heart disease increased for those folks in the study!

[noah]

I think, without over-analyzing the stimulus, (E) simply states that some people with Type A (e.g. 20% risk), became Type B (e.g. 60% risk). Therefore, the risk (probability) for heart disease increased for those folks in the study!

I think this is the heart of the issue.

The first sentence tells us that type B folks are at a higher risk than type A folks - it flat out says that "a person" with type B is at higher risk. So, if you're type B, you're at higher risk. If you've switched from A to B, you've increased your risk. Does it mean those individuals will have heart disease - no. But now, if they were to walk into a hospital, because they're now type B, they'd be assigned a higher risk rating.

I hope that clears it up.