#16- When surveyed about which party they would like to see

[gab1234]

Hello PowerScore friends! I am still confused about C being the correct answer choice. Initially, I chose E, which I understand is incorrect after reading the other posts in the thread; but, I am still stumped on C. How would I come to recognize that C is the correct answer?

[Rachael Wilkenfeld]

The trick to these flaw in the reasoning questions, Gab, is translating the flaw you see to the words they write. It is something that takes slow and careful practice. One of the things I typically recommend is to take the time when reviewing a question to figure out what each of the answer choices is describing, and not just the correct answer choice. For example, on this one, I would note that answer choice (B) is describing a circular answer. Even though that's not the flaw here, the practice in recognizing what the words are describing is really helpful for the test.

So, flaw in the reasoning questions really ask you to do two separate things. First, recognize the flaw in the argument. OFten, students feel that is the ONLY thing they are being asked to do. But it goes a step further. You then have to recognize the description of the flaw you identified in the correct answer choice. And I am sure that I don't need to tell you that they don't always describe the flaws in the most obvious or user friendly ways. Looking at the stimulus, I'd describe this flaw as an error of division. They try to jump from what is true about the whole (the composite results of the survey) to what is true about the parts (each individual person surveyed). When we look at answer choice (C), we see that described as preferences of the group as a whole versus preferences of an individual. That's pretty close to what we recognized in the stimulus. But the more that you practice these, and see the way they are phrased, the easier they will become. You'll notice words like "whole," "group." and "total population" to represent the idea of whole group. The more you do, the more you'll recognize the frequent terms used.

Hope that helps!
Rachael

[VamosRafa19]

I'm not seeing a connection between C and the conclusion. For C to apply, shouldn't the conclusion say "most respondents would like to see"? How are most citizens a individual members of a group that is made up of only respondents?

[Adam Tyson]

Think of that another way - it's saying that most of the individuals who responded to the survey would want to see that combination in the legislature. But isn't it possible that NO individuals want that, even though the group as a whole came out that way?

Respondents = individuals who responded to the survey.

[gingerale]

I recognized that the flaw in the stimulus was an error of division but I was thrown by (C) because I thought it would be the reverse–that the argument takes for granted that the preferences of part of a group are the preferences of the group as a whole. What I gathered from the phrasing was that a number of respondents were surveyed and that it is possible that there could have been any number of respondents: 10, 50, 100, etc. The error I immediately noticed was that the author was drawing a conclusion about the preferences of all citizens based on a survey that very well may have been based on the preferences of a very small fraction of citizens.

[Jeremy Press]

Hi gingerale,

I'm not sure you've put your finger on the precise nature of the "Whole to Part" flaw here. The premise in the stimulus gives us the overall breakdown of the whole public's opinion. In other words, when describing how the "whole" citizenry's opinion breaks down, 40% of that whole want Conservatives, 40% want Liberals, and 20% want Moderates. But that overall breakdown of the results of surveying the whole citizenry doesn't tell me what any single person, or what "most" of the individual persons, within that citizenry are going to want. And it doesn't give me a basis for concluding (as the argument does) that most individual persons are going to want a legislature "split" between 40% C, 40%L, and 20%M. If anything I would expect that 40% of the individual citizens in the country want a legislature composed of ALL Conservatives, 40% of the individual citizens want a legislature composed of ALL Liberals, and 20% of the individual citizens want a legislature composed of ALL Moderates. Under that theoretically possible breakdown, there would be no individual citizen, and definitely not "most" individual citizens, who wanted a "split legislature" (40% L, 40% C, 20% M).

I think you might be getting hung up on the use of the word "most" in the conclusion of the stimulus, which is pretty cleverly disguising the fact that the basic claim the conclusion makes is about individual citizens (who are themselves only parts of the citizenry as a whole).

Let me know if that clears it up!

[zsg2@cornell.edu]

Hello,

Although it wasn't an answer choice, could it also be the case that a flaw in the reasoning is that the author is assuming that the results of the survey is representative of "most citizens" or is that avoided with the clause "if the survey results are reliable"? Thanks.

[Poonam Agrawal]

Hi zsg2!

You're right in thinking that the "representative" flaw is avoided with the the inclusion of, "if the survey results are reliable." The conclusion stands only if that condition is met - that in fact, the survey results are reliable (and therefore, representative).

If the survey results were not representative, then they wouldn't be a reliable measure of most people's opinions, and the argument would break down on its own. Therefore, the author doesn't erroneously make the assumption that the survey results are representative of most people. He/she just says that IF the results are representative, then this is the conclusion we would draw from them. The bigger flaw at play here is the whole-to-part division explained in previous posts. Hope this helps!

[pmuffley]

Hello - I have a similar question to gingerale on this one.

Stimulus: When surveyed about which party they would like to see in the legislature, 40 percent of respondents said Conservative, 20 percent said Moderate, and 40 percent said Liberal. If the survey results are reliable, we can conclude that most citizens would like to see a legislature that is roughly 40 percent Conservative, 20 percent Moderate, and 40 percent Liberal.

40:20:40 was the split among respondents. Let's say we have 10 respondents. That would be a 4:2:4 split.

Then, the stim moves on to say that "we can conclude that most citizens...". Number wise (let's just use America as an example), that would be just over 175,000,000 people. How can we conclude that because 10 people gave their preference, that half of an entire nation would be split the same way. I think that is what gingerale was getting at and where I am stuck as well.

I didn't chose C because whole to part, in my mind, would look like, "Most citizens want a 40:20:40 split. If the survey results giving us those numbers are accurate, then we can determine that my ten friends would want the same thing if asked".

I think I get what you're saying...that the survey group counts as the "whole" and "most citizens" counts as the "part"...but how can that be? Again, a survey group of 10 people would be part of a population of "most citizens".

Please help! Thank you!

[Robert Carroll]

pmuffley,

As always, if something is a flaw in an argument, then fixing the flaw should help fix the argument. So imagine the stimulus did indeed survey everyone on the US. 40% wanted a Conservative legislature, 20% a Moderate one, and 40% a Liberal one. If there is a representativeness flaw in the original argument, that should disappear here. So, in this new situation, do a majority of people want a legislature that is 40% Conservative, 20% Moderate, and 40% Liberal?

Well...no. It's quite possible, even likely, that no one wants that legislature. Ask each of the several hundred million people - do you want a legislature that's 40% Conservative, 20% Moderate, 40% Liberal? The Conservatives could all well say "No, that's too liberal - I want a legislature that's 100% Conservative." The same could happen for the other respondents in the other groups. So no problem has been solved with the argument by making it perfectly representative, and making the sample as large as possible. So no such problem existed.

What the failed fix proves is that what the argument is doing wrong is acting like there's any individual that actually wants the 40/20/40 split. It's like saying "America doesn't like using metric units". Well, America isn't a person with agency - America can't like or dislike anything. What does happen is that individual Americans have preferences, some related to units of measure, and maybe most Americans, most of the time, use inches instead of centimeters. We have to be careful with a loose way of speaking, and the author in this stimulus wasn't. To say, loosely, that "the country wants a 40/20/40 legislature" is perhaps a fair assessment of the data. To say "most people want a 40/20/40 legislature" is not making a claim about the whole but instead about a majority of the individual parts, who are individual citizens. Maybe "the country" wants something, but no actual person, much less a majority of people, wants that thing.

The whole is the country. The parts are the people. The 10-people survey group is NOT the relevant part in this situation. Hope this clarifies!

Robert Carroll