#9 - In a recent field study of prairie plants, the more

[Administrator]

Please post below with any questions!

[Zierra28]

I get the causation/correlation effect, and got this question right. I just want to full eliminate D (the unrepresentative data) because they didn't specify how many prairie plants they used. Is this why the answer is invalid?? When they don't MENTION the number studied, should I generally disregard an answer choice BASED on the "unrepresentative data" possibility?

[Jonathan Evans]

Hi, Zierra,

Good job getting this question right. This question asks you to identify a flaw in the reasoning, and as such your job is to work through the argument to arrive at your own prephrase, i.e. your own description of what you want to see in the credited response.

Proceeding from the stimulus, you should identify the conclusion, the premises offered in support, and in this case give an accurate description of the problem here. For problems such as this, your process might be as simple as asking yourself whether you concur with the author's reasoning. Can you envision any alternate possibilities that are consistent with the facts given?

For instance, I might imagine that there are actually so many different kinds of plants here because the soil is itself very fertile, not because of the plants.

Then turn this description into an abstract prephrase, preferably one that corresponds to one of the common flaws you have studied. In this case something like:

"The author mistakenly concludes that one thing (lots of plants) caused another thing (prairie ability to support life) when the cause could be reversed."

I know that you understand this, but it is essential to your success that you get the process right. Here it appears as though you might be working backwards from the answer choices. Sometimes you'll get the right answer anyways, but you will often get caught in an attractive trap, as you almost did here.

There are two ways you can know D is incorrect:

1) This answer choice describes a survey or sample flaw, which bears little resemblance to the reasoning here.

2) The general conclusion is in fact based on representative data, evidently enough to observe a direct correlation between number of plant species and plant health in different prairies.

To answer your last question, no you should not disregard an answer choice that refers to unrepresentative data based solely on the fact that a number is not mentioned. In fact, you will likely observe that there are many unrepresentative sample flaws that mention no precise numbers. c.f. PrepTest 16, LR Section 3, Problem 11 for an example.

[gcs4v333]

I totally see why (A) is correct, but I'm still having a hard time ruling out (D). For my prephrase, I fixated on the fact that the study focused on a prairie plot ("...the more plant species a prairie plot had..."), but draws a conclusion about an entire prairie ("Thus, having more plant species improves a prairie's ability to support plant life.") It seemed like there was no way that one prairie plot could give you enough information to determine how an entire prairie would react to more plant life.

But as I went through the answer choices I kept (A) because I saw how perhaps the cause/effect was wrong. But I chose (D) because it matched my prephrase better. Between the two, it felt like (A) describes a criticism, but that (D) described the criticism that the argument was MOST vulnerable to.

So how, exactly, could I have seen that (A) is a stronger criticism than (D), when both seem like strong criticisms? Thanks!

[James Finch]

Hi GCS,

The differences between (A) and (D) are that (A) is talking about the possibility of reverse causation, which I always tell students to look for first whenever they identify causal reasoning, as it is very often the correct answer choice when it is logically possible. Most causal relationships wouldn't allow for reverse causation (it would be absurd), so whenever it's a realistic possibility in a stimulus, the question writer likely was thinking about creating a scenario where it was possible in order to test the concept.

Conversely, (D) describes a sampling issue, where the sample used is unrepresentative of the group that the conclusion is making its claim about. In order for that to be the correct answer, we would need to know, by inference, that the experimental group isn't representative of the overall group. Here, we're dealing with a "field study" of prairie plants, and a general claim about the amount of plant species determining how well a prairie can support plant life. Since we don't know anything about the study, we can't say whether the sample is or isn't unrepresentative, so we have no evidence to support (D), while it would be commonsensical to draw the reverse of the conclusion in the stimulus: that the ability of a prairie to support plant life will determine how many plant species it contains, as (A) would indicate.

Hope this clears things up!

[ShannonOh22]

Hi guys,

I see this question's flaw is in the reversal of the possible cause and effect, but I had difficulty due to the odd/different way the flaw is described in Answer Choice A, and (more worryingly) I truly believed C was the right answer. Does the author not commit the part to whole flaw in this argument when he assumes that what happened in one particular prairie plot applies to all prairies? Or is there something specific in the way Answer Choice C is worded that should have tipped me off to the fact that it is incorrect?

Also, I ruled A out because the stimulus seems to deal with three phenomena...more plant species in the plot = X, the more vigorously the plants grow = Y, and the better the soil retained nutrients = Z. Clearly here Z was added on my part. Was my mistake in separating the vigorous growth from the soil nutrient retention? Those seem to be two different things to me, but should I assume they are part of the same "phenomena" for the purposes of this argument? What exactly should we understand the word "phenomenon" to mean in LSAT speak? Phenomenon has multiple definitions in the real world - simply an observable event, something pertaining to nature, something that is impressive or extraordinary...is there one definition we can universally depend upon for the purposes of the LSAT?

Thanks in advance for your help and insight!

[Paul Marsh]

Hi Shannon!

The trouble with answer choice (C) is that the stimulus wasn't talking about just one plot. Instead, it was relying on an entire field study that appears to have surveyed multiple plots ("the more plant species a prairie plot had..."). If the first sentence of the stimulus had said something akin to, "In studying the prairie plot behind my house, I noticed over time that the more plant species the plot had, the more vigorously the plants grew and the better the soil retained nutrients," then (C) probably would have been a good answer. As it is, however, the stimulus never relied on just one plot to make its point.

For your second question, I wouldn't say it's a mistake on your part to think of "vigorously growing" as Y and "soil retaining nutrients better" as Z; they are, as you say, "separate phenomena". But the important thing to key in on here is not whether Y and Z are separate phenomena, but the fact that the author says X caused Y and Z. The use of causal reasoning in any argument is inherently a gap in that argument. Whenever you recognize the use of causal reasoning in a Flaw question (or certain other question types like Weaken), alarm bells should be going off in your head. At that point, it's time to look for an answer choice that says something like "The argument ignores an alternate cause" or, like here, "The argument infers that X causes Y without considering the opposite." Identifying the use of causal reasoning in an argument and familiarizing yourself with the typical causal reasoning Flaw answer choices are the two take-aways from this question that will serve you well on future problems; thinking of "vigorously growing" and "soil retaining nutrients better" as separate events is fine but don't let it distract you from those two main points. Hope that helps!

[Pinkxo]

I choose answer D - bases a general conclusion on data that is likely to be unrepresentative.

I thought this was the answer because it just mentions ONE study, so my thought process here was how can one study on prairie plants show us it improves the prairie’s ability to support plant life.
Seems like it was generalizing from one study but I think I went wrong here because there’s not enough information about the study, what if the sample is representative and this study did like 5,000 tests and found in all these tests more plant specifies a prairie plot had the more vigorously the plant grew and retained soil.

Is that the reason answer choice D is incorrect?

[Robert Carroll]

Pinkxo,

You're right on this one. Imagine that answer choice (D) were the flaw. Then the problem with the stimulus is that the data might not be representative. If that's the problem, then let's fix it! Imagine someone responds to our criticism by conducting 5000 more studies, making sure that the aggregate is representative. Every study shows the same results. If the flaw were what answer choice (D) said, then this course of action should solve the problem with the stimulus. But it doesn't - in every of those 5001 studies, it's possible that the correlation between number of species and plant vigor might not be due to the causal connection (more species produces a prairie better able to support plant life) in the conclusion. It might be that the correlation is due to other factors, that the correlation can be explained by reversing cause and effect, and so on. So, while the study might not have been representative, that's not the flaw with the argument - the flaw is that, even if representativeness were proved, the causal claim in the conclusion would still not be established.

Robert Carroll