#11 - Recent studies indicate a correlation between damage

[Administrator]

Complete Question Explanation

Flaw in the Reasoning. The correct answer choice is (A)

The author of this stimulus draws a questionable causal conclusion based on two very limited
premises: Studies show a correlation in humans between damage to chromosome number six and
schizophrenia. But, the author points out that there are people who have the damaged chromosome
without the disease, and there are people who have the disease without any damage to the
chromosome. All that this actually proves is that there is not a perfect correlation between the two
conditions, but the author incorrectly jumps to the conclusion that there can be no causal connection
between the two.

The vagueness of the wording chosen by the author provides only that there are at least “some” (this
might mean only one or two!) who have a damaged chromosome without having schizophrenia, and
there are “some” who have the disease without damage to the chromosome. The existence of such
people (who might just be exceptions to a general rule) does not prove that there is no causal link
between the disease and chromosome number six—the disease and the chromosome may be very
closely correlated, even if there do exist some people who have one without the other (even if there is
not a perfect correlation).

The stimulus is followed by a Flaw in the Reasoning question, which should not be surprising,
considering the weak argument presented in the stimulus. The correct answer choice will probably
point out that the author incorrectly takes the existence of “some” exceptions as disproof of any
causal relationship.

Answer choice (A): This is the correct answer choice. The author clearly believes that a causal
link would require perfect overlap of the two attributes (the disease and the damaged chromosome).
This answer presents another way to phrase this point: the author ignores the possibility that not
every type of chromosome damage leads to schizophrenia—that there might be a causal relationship
even if the two conditions will not always be linked

Answer choice (B): The author does not presume that chromosomal damage is the only cause of
schizophrenia; the flaw in this stimulus is the author’s conclusion that there is no causal link, just
because there is not a perfect correlation.

Answer choice (C): There is no sample population discussed in this stimulus; the flaw here is not an
unrepresentative sample. The author points to certain specific exceptions to the correlation between
schizophrenia and a damaged chromosome (people who have one but not the other), but no sample
population is presented or referenced.

Answer choice (D): While the flaw found in this stimulus does involve causal reasoning, the author
does not mistake a cause for an effect. In fact, the author never specifies which of the two is believed
to be the cause and which is believed to be the effect—only that there is not a perfect correlation
between the two.

Answer choice (E): The author does not presume that correlation implies causation; in fact, this
incorrect choice is very nearly an Opposite Answer, because the actual flaw in the author’s reasoning
is the mistaken presumption that any relationship short of perfect correlation cannot be a causal
relationship. In reality, of course, there can be exceptions to general causal relationships.

[jared.xu]

Hello,

I have a question from the Take-Home Test September 2009. LG 1 problem #11 (11. Recent studies indicate a correlation between damage to...). I thought this was a classic "causal" problem that Powerscore has taught us, and that one of the five ways of destroying the causal relations that Powerscore has taught us would apply. That is, a causal relation is invalid if we are shown that the cause happens without the effect or that the effect happens without the cause. The stimulus of this problem shows exactly both cases. And so I eliminated A because it suggests the possibility of a causal relationship despite violating two of Powerscore's five ways of destroying the causal relations. However, A ended up being the answer. I arrived at E also because of the violation. The stimulus concludes that there is no causal connection. And if we apply Powerscore's five ways, there would not be any logical gaps. Yet, the question stipulates that we have to describe a flaw. I chose E because it suggests that there is no causal relationship in the first place, and that the conclusion is wrong to assume people feel that the correlation implies a causation. Could you please tell us why A is right, and E is wrong? And how does this problem not violate what we were taught by Powerscore's five ways of destroying causal relations? And finally, how can we prevent in the future this problem of getting a question wrong precisely because we faithfully applied the Powerscore technique of destroying causal relations? Thank you in advance for replying.

[Dave Killoran]

Hi Jared,

Thanks for the question. Let’s take a look at the problem you reference, because you appear to have missed several critical factors in the question.

First, and most importantly, the “five ways of destroying the causal relations” refers to Weaken question answer choices as presented by LSAC. The question you reference is actually a Flaw in the Reasoning question, which asks you to describe what occurred in the stimulus, not to weaken the answer with an answer choice. So, in this instance you aren’t attempting to affect the argument (as in a Weaken question) but rather to describe what the author did (Flaw in the Reasoning ). Those two acts are fundamentally different, and have a lot to do with why you missed this question.

Second, the conclusion in the problem you reference is that there is “no causal connection.” So, you aren’t dealing with the classic causal scenario we describe in Weaken questions, which is where the author makes an error of assuming that one thing causes another. As you might suspect, that changes things a bit.

You ask “how can we prevent in the future this problem of getting a question wrong precisely because we faithfully applied the Powerscore technique of destroying causal relations.” From the above two points, you can probably see that the key is to apply the right ideas at the right time. If you use the right tool at the right time, your chances for success are greatly increased. If you pull out the wrong technique, anything can happen

Let's take a look at the two answer choices you asked about:

Answer choice (A): This is the correct answer. The premise of the argument clearly posits a correlation. Is it possible that there is causality here? Of course--there are numerous LSAT problems where causal relations are discussed as fact. So, where does the error in causality occur (when it occurs) in those problems? It occurs when someone blithely concludes that causality is present (the mistake being that you can't be certain). Here, a correlation is present, and on the basis of some contrary information, the author blithely concludes that causality is definitely not present. This is the same type of assumption, but in the other direction (the mistake being that, again, you can't be certain). The error here is really drawing a conclusion on the basis of inadequate information, and (A) describes that in terms more specific to this problem.

Answer choice (E): This answer describes the exact opposite of what the author does. The author states that recent studies indicate a correlation, and then concludes that there is no causal connection. This answer states that the author presumes that a correlation implies causation--the exact opposite of what occurred.

Please let me know if the answer above helps. Thanks!

[jared.xu]

Hi Dave,

Thank you so much for your reply. It clears up my misunderstanding of what entails causal relations. So is it true to say that the "five ways of weakening causal relations" apply only to causal scenarios in Weaken questions? It appears to be more useful to think of "the five ways of weakening and strengthening causal relations" as part of a bigger picture that reveals the assumption the author makes. So is it true that if we encounter a causal scenario, we shouldn't immediately start applying the "five ways," but should rather wait to see what the assumption is? By focusing on the assumption rather than the "five ways" during a causal scenario, I hope I could apply the right ideas at the right time. (I am currently listening to Jon Denning's "Question Type Relationships" virtual module.) I am also a big fan of your virtual modules. You've been a great help to me not just this once. I really appreciate it.

Gratefully,
Jared

[Dave Killoran]

Hey Jared,

Thanks for the kind words--I really appreciate them! I'm really glad I've been able to assist you so far

Let's talk a bit about the questions you ask. The five ways of weakening causal scenarios are based on an analysis of LSAT Weaken question answer choices over the years. Those five ways appear consistently in that type of question. Can the test makers use those five ways to weaken their own argument? Sure, but it doesn't happen that often (and it would appear in a context where either two speakers were used in the stimulus, or, less likely, where an author clearly delineates the positions of two different groups on an issue).

So, what should happen when we see causality discussed in a stimulus? First, it depends on where the causality occurs. Remember that when causality is discussed as part of the premises, the test makers typically accept it as fact, and move on from there. On the other hand, when we see causality in the conclusion (as in, a causal conclusion based on non-causal premises), that's when we know that the author is making the assumption that the cause makes the effect occur (or, if the conclusion denies that a causal relationship occurs, that the author assumes that there is no cause and effect relationship present). Second, it depends on what type of question stem we see. If it's a Weaken question, then yes, I start analyzing answer choices to see if they match the five ways. If it's a Strengthen question, then I look for the opposite of the five ways. If it's a Flaw question, I look for a description of what occurred, etc. The analogy that I like to use is that you are like an air-traffic controller--you have to recognize what pops up on the screen, track it, and then deal with it appropriately.

So, to answer your question directly ("is it true that if we encounter a causal scenario, we shouldn't immediately start applying the "five ways"), the answer is that you should wait until you see what the test makers have given you, both in the stimulus and in the question stem. The second part of your question ("but should rather wait to see what the assumption is") is also interesting. In classic causal problems, the assumption is literally that the cause always makes the effect happen. One of the issues you ran into in the question you originally asked about what that it looks like you saw the word "correlation" and then immediately went into causal weakening mode. That proved problematic because the author actually made a judgment that no causality was present, and then they asked a Flaw question. The point is that you have to see what the author does before you start applying a certain form of analysis. In the Question Type Relationships module, what you are seeing is a discussion of classic combinations that appear frequently. But, note that word "frequently," because that means they often happen together, not that they always have to happen together. The people who make this test are very, very good at what they do and on occasion they will go in a different direction than expected. Again, it is your job in those situations to play air-traffic controller and note what they do. The good news is that this task gets easier over time as you see different variations on the concepts. The more questions you see, the more you will begin to understand what they like to do and how they do it.

Please let me know if that helps. Thanks!

[jared.xu]

Hi Dave,

Thank you so much for your that lengthy explanation. It would be nice to have another quick confirmation from you. When you state in your last paragraph of explanations that "in classic causal problems, the assumption is literally that the cause always makes the effect happen," do you mean to invoke the five ways of supporting a causal conclusion? If that is the case, could we add to your statement by saying that in classic causal problems, the assumption is literally that the cause and nothing else always makes the effect happen, and that the effect could only happen with the cause preceding it, and that the data used to make the causal statement is accurate? I am trying to make sure I understand what you mean by that sentence.

Thank you,
Jared

[Dave Killoran]

Hey Jared,

I wasn't directly trying to invoke the five ways of supporting causality. On the LSAT, think of those as answer choices that provide additional support for an already-existing causal conclusion. Keep in mind, though, that based on the way the argument is made, an LSAT author thinks those are already in operation (just as he or she assumes that the ways of weakening the argument don't exist).

In LSAT causality, the classic assumption is that the cause always makes the effect happen. Could other elements make the effect happen? Sure. For example, consider a conclusion that states, "skipping the final exam caused Jerry to fail the class." The LSAT could easily allow other causes to exist--for example, not studying for the class. The key is, where does that consideration appear? If it appears in the stimulus, then you simply consider it as part of the overall information sphere, and proceed based on the statements they make.

The question that naturally follows from that, is why would a statement like "Jerry failed to study for the class" then weaken the argument? The answer is this: because, unless stated otherwise, in the argument the author just assumed that in this instance skipping the class was the one and only cause of failure (otherwise, he or she would have mentioned other causes). Thus, raising any alternative weakens the argument. This is a tricky point, but at the same time a pretty cool one.

From our discussion, I sense that you want to keep putting the views ways into the conclusion. That's not how they are designed to work--they are answer choice models that follow from the basic assumption inside a causal conclusion.

The key is to not attempt to view causality (or any reasoning form) in simplistic terms. As much as I would like to reduce it to a series of binding black and white statements, it's really too messy to do that. We can reduce parts of it, but there are often gray areas in reasoning, and the test makers always have the option of exploiting the situation by changing some of the variables.

Please let me know if that helps. Thanks!

[jared.xu]

Hi Dave,

Thank you so much. Your explanations are all very helpful. But they give me a further desire to ask you to confirm whether my current understanding is correct, at the risk of appearing too simplistic to you. When you talk of the classic assumption of LSAT causality, are you saying that in an assumption question where a causal relationship is established in the conclusion, the correct answer to this assumption question will very likely be something like “the cause always makes the effect happen,” or at the very least, the idea (“the cause always makes the effect happen") will always be an implicit assumption to this assumption question? Are you also saying in your last post that if an alternative cause is not mentioned in the stimulus, the correct answer to this assumption question could also be that another alternative cause does not exist? I know these two questions make it very hard to answer without concrete scenarios, so thank you in advance for trying.

Always gratefully,
Jared

[Dave Killoran]

Hey Jared,

No problem--I don't think you are being too simplistic, but I sometimes get cautious when an idea as big as causality is approached with the idea that it can be rigidly defined. In this case, I don't think these questions are limiting at all

Let's break your question down:

When you talk of the classic assumption of LSAT causality, are you saying that in an assumption question where a causal relationship is established in the conclusion, the correct answer to this assumption question will very likely be something like “the cause always makes the effect happen,” or at the very least, the idea (“the cause always makes the effect happen") will always be an implicit assumption to this assumption question?

The correct answer will either confirm/reinforce the cause/effect relationship, or eliminate possible challenges to the cause/effect relationship. This is in line with the Supporter/Defender Assumption Model we use to explain the role of assumptions. The author does assume that the cause always makes the effect happen (in the scenario you describe). Whether or not that will be the answer is up to the test makers.

Are you also saying in your last post that if an alternative cause is not mentioned in the stimulus, the correct answer to this assumption question could also be that another alternative cause does not exist?

Yes, definitely!

Please let me know if that helps. Thanks!

[jared.xu]

Hi Dave,

Thank you for all your comments. I apologize that our discussion took a very theoretical and abstract turn. But it is really useful for me to have this general understanding of the workings of causal relationships on the LSAT. Really appreciate your patience and guidance!

Jared