#19 - Viewers surveyed immediately after the televised

[Administrator]

Complete Question Explanation

Weaken—CE. The correct answer choice is (D)

This stimulus presents a causal conclusion in which the stimulus author chooses (poorly) between two potential causes for a candidate’s election win.

Last year there was a televised political debate between Lopez and Tanner. Viewers surveyed immediately after the debate tended to think that Lopez had made the better argument, and Lopez later won the election. The stimulus author treats the fact that Lopez won the election as evidence that the survey respondents who reported favoring Lopez’s debate arguments did so because they were biased in Lopez’s favor.

This conclusion is causal and, like all causal conclusions in the Logical Reasoning section, it is flawed. The author considers the bias in Lopez’s favor to be a cause that led to at least two results: the survey responses and the election result.

• C E
  
  survey result
  
  bias +
  
  election result

However, it may be that the survey result led to the election result, or that some other cause, such as Lopez’s talent, that led to both the survey and the election results. In fact, it likely was a mixture of various elements that led to Lopez’s victory. The prephrase in this Weaken question is that the correct answer choice will attack the causal conclusion by showing that bias in favor of Lopez may not have produced the survey results favoring Lopez. Move to the answer choices with this prephrase, and do not spend time trying to predict precisely how the correct answer choice will accomplish this.

Answer choice (A): The fact that most people who voted in the election did not watch the debate is irrelevant to the conclusion, because the conclusion was limited to the issue of whether some of those who did watch the debate were biased in favor of Lopez.

Answer choice (B): This information is not relevant to the conclusion, which considered only that those respondents who reported that Lopez’s arguments were more persuasive may have been biased in favor of Lopez.

Answer choice (C): This answer choice is incorrect, because the likelihood of those who watched the televised debate to vote for Tanner rather than Lopez is not relevant to the question of whether those respondents who stated that Lopez was more persuasive responded the way they did because of bias.

Answer choice (D): This is the correct answer choice. The evidence in the stimulus established that the debate viewers surveyed immediately after the debate tended to think Lopez had made better arguments. The argument concludes that those who responded favorably for Lopez did so because they were biased in favor of Lopez. However, if it is the case that most of the viewers surveyed prior to the debate said they would probably vote for Tanner, Lopez’s opponent, then this casts some doubt on the conclusion that the survey respondents answered the way they did because of a bias toward Lopez. This answer choice does not prove that the respondents at issue were not biased toward Lopez, but it does cast some doubt on the conclusion.

Answer choice (E): The margin of Lopez’s victory has no relevance to the question of whether the survey respondents who said Lopez’s arguments were better during the televised debate were biased in his favor.

[jwheeler]

I chose D because it was clearly the best answer, but I was a little concerned about one detail. It never specifies that the people surveyed before were the same ones they surveyed after. So it could've been the case that they didn't have a representative sample before the debate and in fact, more people were planning to vote for Lopez from the onset. So their bias still factored into their response to the survey. Is there an assumption that the survey results wouldn't have been terribly unrepresentative? Or am I maybe being too cautious about this, as the testmakers sometimes like to use survey misinformation in their own questions.

[Brook Miscoski]

jwheeler,

It is true that you could get into the weeds on a number of issues. Did Tanner supporters have an earlier bedtime, and go to sleep rather than responding to the post-debate survey? Is it actually necessary to avoid re-surveying initial respondents because people who have already given one answer may be unwilling to change it, and is your group size big enough that eliminating them from the second survey isn't a problem? But barring some weirdness or technicality that requires statistical training, it doesn't matter whether the same people took the first survey as took the second survey. A couple random surveys of the same group (viewers) at different times should describe the group attitudes at different times.

So yeah, on this one just don't overthink it. It's a weaken question, and this answer choice gives you a strong reason to think viewers weren't biased, they were won over.

[hope]

Even with your explanation, I still don't get it. In A, B, C, you state that these answers are outside of the persons interviewed immediately after debate. Answer D contradicts this statement because it, too, is outside the persons interviewed immediately AFTER the debate. Applying your logic, Answer D is wrong because its respondents were interviewed PRIOR TO the debate. It seems to me that Answer A should be the correct answer because if they voted for Lopez without even seeing the debate, that proves that they did not favor Lopez and objectively cast their vote for Lopez with no sense of bias. This proves that bias did not necessarily have to be the cause for the election results. Please respond. Thank you.

[hope]

I just had an epiphany. If in fact the persons who were interviewed PRIOR TO the debate were at least some of the same persons interviewed AFTER the debate, then clearly this change of mind does seem to point to a lack of bias. This change of mind can point to the fact that while favoring Tanner before the debate, Lopez's excellent debate performance literally prompted the change of mind. Am I right in thinking this way? Thanks for your response.

[Claire Horan]

Hi Hope,

You're on the right track, but I agree with Brook that it doesn't matter whether they interviewed the same people or different people. As she explained, "A couple random surveys of the same group (viewers) at different times should describe the group attitudes at different times." Regardless, it sounds like you get the idea.

Answer choice D suggests that the group's attitude toward Lopez became more favorable as a result of the debate. That suggests that Lopez was effective in the debate, which weakens the author's conclusion that bias motivated the survey respondents to report that Lopez's arguments were better.

Nice job thinking through this one!

[hope]

Thanks for the explanation Claire. I love Powerscore for that very reason.

[plzhelpme101]

Can someone please explain how C is incorrect? If you watched the debate and then said that you were more likely to vote for Tanner, would that not blow up the conclusion that those who responded to the survey were biased in favor of Lopez? Unless the survey respondents included people who did not watch the debate, but that is reaching, even by LSAT standards.

I’m so confused

[Jeremy Press]

Hi plzhelpme,

It sounds like you might be misreading answer choice C. You said, "If you watched the debate and then said that you were more likely to vote for Tanner...," but that isn't what answer choice C is describing. Answer choice C is not necessarily talking about what people said when they were surveyed after the debate. Instead, answer choice C is just talking about the general voting preferences of viewers: "people who watched the televised debate were more likely to vote for Tanner than were the people who did not watch." In other words, answer choice C doesn't specify at what point in time those people had or developed that preference for Tanner. Further, answer choice C is comparing them to an irrelevant group (people who did not watch the debate). The group of people who did not watch the debate doesn't shed any light on whether the people who did watch were biased for Lopez or not.

I hope this helps!

[Sufficient_Necessary]

I found that the key to this question was being really clear about the conclusion at issue: the survey is trying to assess whether viewers thought Lopez or Tanner won the debate, and the author of the stimulus is arguing with the conclusions of the survey. The survey suggests viewers thought Lopez won. The author isn't convinced.

With that in mind, I'd add this to the discussion on Answer Choice (C):

Answer Choice (C), it it does anything, actually strengthens the argument of the author of the stimulus. If it's the case that viewers of the debate were more likely to vote for Tanner after the debate, then it is definitionally the case that the sample of viewers taken after the debate is not representative of viewers of the debate as a whole. If this answer choice is true, the author might very well be right that the sample *was* biased -- the post-debate opinions of the sample of viewers (pro Lopez) don't appear to have reflected the post-debate opinions of overall population of debate viewers (pro Tanner). Having said that, similar to other answer choices, this answer choice is ultimately about voting and not about opinions on debate winning. There are plenty of reasons voters might believe Lopez won and still vote for Tanner, leaving ample room for the possibility that the survey accurately captured the opinions of debate-viewers: they thought Lopez won. But they voted for Tanner anyway. So this answer choice doesn't necessarily say anything about debate-winning opinions at all.

(A): Answer choice (A) can be true and not affect the strength of the argument in the stimulus. The issue at hand is whether viewers thought Lopez or Tanner won the debate. Pointing out that over 50% of people that voted in the election didn't watch the debate doesn't tell us anything about what the people who did watch the debate thought about the debate they watched.

(B): Similarly, answer choice (B) is about the live audience. The live audience is not representative of viewers that watched the debate on television. In the most literal reading of the stimulus, the point-at-issue is which candidate the audience that watched the televised debate thinks won; answer choice (B) tells us about what viewers that watched the debate live (not on television) thought -- so it's about the a totally different set of people; the wrong people for our purposes. In the loosest reading of the stimulus, the viewers of the "televised debate" includes both in-person and television viewers. In which case the in-person viewers aren't irrelevant, just unrepresentative -- they aren't a good measure of which candidate all viewers thought won the debate.

(E) Answer choice E, like (A) and (B) is now dealing with the winner of the election, not whether viewers of the debate thought Lopez won the debate. And in particular, it's not dealing with whether the viewer-sample that thought Lopez won is representative of the larger pool of people that watched the debate. The only connection between Answer Choice (E) and the stimulus, if there is one at all, is that the stimulus mentions as proof of its claim that Lopez won the election. In this more indulgent reading, answer choice (E) is nevertheless consistent with the stimulus. And an answer choice consistent with the argument can't weaken the argument. The author of the stimulus doesn't say that the survey was a landslide in favor of Lopez, only that the viewers *tended* to think Lopez won the debate... and that they *may* have been biased. We can believe answer choice (E) to be true and still adopt the fallacious logic of the author: we can easily read the argument to mean that Lopez narrowly won the debate, went on to narrowly win the election, and therefore the sample was narrowly biased for Lopez. If we can believe the answer choice and the flawed logic of the stimulus together, the answer choice doesn't weaken the stimulus.

The straining we have to do to see the connection between the election results and bias in a survey about debate results speaks to the many other more obvious answers for why a survey about who won a debate might line up with who won the election. Namely, that Lopez won the debate, and winning the debate was predictive of winning the election; that the survey of viewers showed that a majority of respondents thought Lopez won because Lopez did a good job, and that the survey after the debate was more/less representative of what all viewers thought, as witnessed by the ultimate election results. If it's the case that debate viewers surveyed before the debate said they were more likely to vote for Tanner... that's not helpful or consistent with the author's claim. That now requires explanation just to get back to neutral. As folks above have point out, the author might argue, for example, that the people surveyed weren't the same people as the post debate survey -- but it's not clear why that would matter. We don't need to disprove the author's retort to challenge presented by the answer choice, it's enough that the answer choice would knock the author would be back on their heals. If (D) is true, the fact that the survey results flipped from one survey to the next is not dispositive -- it doesn't break the author's argument beyond all repair, but it's definitely unhelpful. It raises doubts -- and that's all the question asks us to do.