probability of vote and issue preference

[gupviv23]

Hi

I am not able to understand the graph here. As per my understanding, for first question- most company with most stagnant variation (SIGMA) should be the answer and for second ZETA should be the answer. Can you please explain?
[img=http://www2.picturepush.com/photo/a/8789620/220/8789620.jpg]

Thanks
Vivek

[tim]

the y-axis indicates how likely someone is to vote the same as their stated preference, so if you want to identify the party whose members are most likely to vote the same, you need to find the party with the highest graph - Delta..

i agree with you that the second answer should be Zeta. do you have reason to believe that's not the correct answer?

[gupviv23]

Hi Tim

Thanks for the reply.

Can you please elaborate why we need to look for the company with the highest graph for 1st question?

I think I am not understanding the graph completely, can you also please explain the graph....


Thanks
Vivek

[tim]

i have explained how to interpret the graph and why we need to look for the highest graph. if my explanation didn't help, can you please elaborate more on what you didn't understand so i can help you further?

[gupviv23]

Hi tim

i think i am not understanding the graph properly, please find below my queries on the same-
1) x-axis shows for and again on two points. what other points on x-axis depicts?
2)if we say probability of a company is 0.7 for AGAINST and not it should be 0.3 for FOR?
3) You said, we should look for highest graph- are you looking for highest FOR and AGAINST points? If yes, then what other three points in between those means here?

Thanks
vivek

[tim]

1) x-axis shows for and again on two points. what other points on x-axis depicts?

some level of opinion that is somewhere between being completely for and completely against..

2)if we say probability of a company is 0.7 for AGAINST and not it should be 0.3 for FOR?

if the graph showed .7, that would represent the likelihood that the person would vote the same way as they indicated in their preference. and, although it seems irrelevant here, there would be a .3 probability of voting differently from their preference..

3) You said, we should look for highest graph- are you looking for highest FOR and AGAINST points? If yes, then what other three points in between those means here?

highest overall. as i mentioned earlier, the points along the x-axis are various opinion levels along a continuum from "for" to "against"..

[ivanushk]

First of all, you have to agree that this is a very confusing graph. Second, the way I see it is that ,say, Delta starts with 60% preference "against" and ending with 80% "for" so the way I read it is Delta starts with one side of the spectrum and goes to another - so logically Delta is NOT most apt to vote according to its previously stated preference.

I know I am seeing it wrong, because I know that Delta is the correct answer but perhaps, after reading the post, this is where the problem is. I am willing to learn but this is just...

And frankly speaking, GMAT goes out of its way to mess people up - this is supposed to be "real" life example or reflect "real" life - well, whoever uses this graph may not be understanding what he or she is really looking at him/herself. If I came to a client with this representation of my view and this would be the only graph I had I would probably get fired right there or at the very least the meeting would be over with a phrase "anything else?".

May be if they took away "against" and "for" - it would make more sense but I am still in doubt over the meaning of it the way it looks right now. Sorry Tim (may be you can add more to the above) but it just does not make sense.

[RonPurewal]

First of all, you have to agree that this is a very confusing graph.

if you are confused, you should go back to the introductory paragraph (in the top right corner) and read it more carefully. yes, the wording is fairly dense and awkward, but it tells you everything you need to know.

Second, the way I see it is that ,say, Delta starts with 60% preference "against" and ending with 80% "for" so the way I read it is Delta starts with one side of the spectrum and goes to another

no, there's no time-series on the graph. i.e., you are interpreting the x-axis as though time were elapsing from left to right; that's an incorrect interpretation.
you should never assign time to an x-axis, unless the units on the x-axis actually represent passing time.

for instance, let's say i make a scatterplot of the heights and weights of the players on a football team. if i put the heights on the x-axis and the weights on the y-axis, this most definitely does not imply that the players are getting taller with passing time.

for each initially stated preference (= point on the x-axis), the y-coordinate represents the percentage of people with that initial preference who voted the same way.

it doesn't actually matter whether the stated preferences are "for" and "against". if you don't like those, or find them confusing, here are five new labels for the x-axis:
"wanted black car"
"wanted white car"
"wanted red car"
"wanted blue car"
"wanted gray car"
in this case, if the first y-coordinate is 0.6, that would mean 60% of the people who originally wanted a black car actually bought one. and so on.
the point is that each x-coordinate, on each line, represents a completely different group of people. with the new coordinates above, you have your people in each party who originally wanted black cars; your other people in each party who originally wanted white cars; and so on.

And frankly speaking, GMAT goes out of its way to mess people up - this is supposed to be "real" life example or reflect "real" life - well, whoever uses this graph may not be understanding what he or she is really looking at him/herself. If I came to a client with this representation of my view and this would be the only graph I had I would probably get fired right there or at the very least the meeting would be over with a phrase "anything else?".

well, fine. but they have to write problems that some people actually get wrong!
a beautiful, elegant, easily read, easily interpreted graph would make a horrible standardized-test problem, because everyone and their mother would get it right, and it wouldn't serve to differentiate anyone.

May be if they took away "against" and "for" - it would make more sense but I am still in doubt over the meaning of it the way it looks right now. Sorry Tim (may be you can add more to the above) but it just does not make sense.

hopefully, my analogy with "here's the color of car they originally wanted" and "this many % actually got the color they originally said they wanted" will help a bit. let me know.

[ivanushk]

Thanks a lot about going into details and sorry if it sounded a bit emotional. It makes a little more sense now. Still,

it doesn't actually matter whether the stated preferences are "for" and "against". if you don't like those, or find them confusing, here are five new labels for the x-axis:
"wanted black car"
"wanted white car"
"wanted red car"
"wanted blue car"
"wanted gray car"
in this case, if the first y-coordinate is 0.6, that would mean 60% of the people who originally wanted a black car actually bought one. and so on.
the point is that each x-coordinate, on each line, represents a completely different group of people. with the new coordinates above, you have your people in each party who originally wanted black cars; your other people in each party who originally wanted white cars; and so on.

for and against really makes it difficult to orient in a timed environment. But relating to the above, I had a question: if you say that

in this case, if the first y-coordinate is 0.6, that would mean 60% of the people who originally wanted a black car actually bought one

how to read a) 0.7 (Delta) and b) the change from 0.6 to 0.7? I understand now that 70 percent who wanted something actually did it, but if I assign no time to it, I am getting lost. Or 0.7 can be a "different product" so to speak? Basically, I am not sure how to read the increase or decrease in this graph. Or, now that I read you more carefully, it means that 6 out first batch of, say, 10 people (or 60%), from Delta who originally said "for/against" eventually confirmed it, and 7 out of another batch of 10 (within Delta) did the same? And if Delta had, for example, 50 people then we see the dynamics within those different batches. Is this the right way to interpret this graph?

Thanks in advance.

[RonPurewal]

Or, now that I read you more carefully, it means that 6 out first batch of, say, 10 people (or 60%), from Delta who originally said "for/against" eventually confirmed it, and 7 out of another batch of 10 (within Delta) did the same? And if Delta had, for example, 50 people then we see the dynamics within those different batches. Is this the right way to interpret this graph?

Thanks in advance.

yeah, that's what i get from it.

note that this interpretation is still not 100% airtight. specifically, we don't actually know that these batches are the same size. if the first (left-hand) "batch" is, say, 99% of the whole population, then zeta would beat delta overall.

that's pretty unlikely, though; it's much more likely that the other four batches outnumber the left-hand batch. since the problem just asks us for the "most accurate" statement, we can confidently pick delta.

for and against really makes it difficult to orient in a timed environment.

well, sure -- that's what makes the problem challenging.

[ivanushk]

Many thanks (sorry it took me a while to respond), your response is a great relief! I felt so frustrated about this question.

[tim]

:)

[nyc917]

First of all, you have to agree that this is a very confusing graph.

if you are confused, you should go back to the introductory paragraph (in the top right corner) and read it more carefully. yes, the wording is fairly dense and awkward, but it tells you everything you need to know.

Second, the way I see it is that ,say, Delta starts with 60% preference "against" and ending with 80% "for" so the way I read it is Delta starts with one side of the spectrum and goes to another

no, there's no time-series on the graph. i.e., you are interpreting the x-axis as though time were elapsing from left to right; that's an incorrect interpretation.
you should never assign time to an x-axis, unless the units on the x-axis actually represent passing time.

for instance, let's say i make a scatterplot of the heights and weights of the players on a football team. if i put the heights on the x-axis and the weights on the y-axis, this most definitely does not imply that the players are getting taller with passing time.

for each initially stated preference (= point on the x-axis), the y-coordinate represents the percentage of people with that initial preference who voted the same way.

it doesn't actually matter whether the stated preferences are "for" and "against". if you don't like those, or find them confusing, here are five new labels for the x-axis:
"wanted black car"
"wanted white car"
"wanted red car"
"wanted blue car"
"wanted gray car"
in this case, if the first y-coordinate is 0.6, that would mean 60% of the people who originally wanted a black car actually bought one. and so on.
the point is that each x-coordinate, on each line, represents a completely different group of people. with the new coordinates above, you have your people in each party who originally wanted black cars; your other people in each party who originally wanted white cars; and so on.

And frankly speaking, GMAT goes out of its way to mess people up - this is supposed to be "real" life example or reflect "real" life - well, whoever uses this graph may not be understanding what he or she is really looking at him/herself. If I came to a client with this representation of my view and this would be the only graph I had I would probably get fired right there or at the very least the meeting would be over with a phrase "anything else?".

well, fine. but they have to write problems that some people actually get wrong!
a beautiful, elegant, easily read, easily interpreted graph would make a horrible standardized-test problem, because everyone and their mother would get it right, and it wouldn't serve to differentiate anyone.

May be if they took away "against" and "for" - it would make more sense but I am still in doubt over the meaning of it the way it looks right now. Sorry Tim (may be you can add more to the above) but it just does not make sense.

hopefully, my analogy with "here's the color of car they originally wanted" and "this many % actually got the color they originally said they wanted" will help a bit. let me know.

This problem (question prompt and the graph) gave me a lot a trouble as well.

Your analogy is great, as always. This is why you are awesome.

[tim]

Let us know if there are any further questions on this one.