Two different groups of test-takers received scores on the : PS Archive
Check GMAT Club Decision Tracker for the Latest School Decision Releases http://gmatclub.com/AppTrack

 It is currently 16 Jan 2017, 07:59

### GMAT Club Daily Prep

#### Thank you for using the timer - this advanced tool can estimate your performance and suggest more practice questions. We have subscribed you to Daily Prep Questions via email.

Customized
for You

we will pick new questions that match your level based on your Timer History

Track

every week, we’ll send you an estimated GMAT score based on your performance

Practice
Pays

we will pick new questions that match your level based on your Timer History

# Events & Promotions

###### Events & Promotions in June
Open Detailed Calendar

# Two different groups of test-takers received scores on the

Author Message
Director
Joined: 07 Nov 2004
Posts: 689
Followers: 6

Kudos [?]: 142 [0], given: 0

### Show Tags

30 Nov 2004, 09:38
00:00

Difficulty:

(N/A)

Question Stats:

0% (00:00) correct 0% (00:00) wrong based on 0 sessions

### HideShow timer Statistics

This topic is locked. If you want to discuss this question please re-post it in the respective forum.

Two different groups of test-takers received scores on the GXYZ standardized test. Group A's scores had a normal distribution with a mean of 460 and a standard deviation of 20. Group B's scores had a normal distribution with a mean of 520 and a standard deviation of 40. If each group has the same number of test-takers, what fraction of the test-takers who scored below 440 belonged to Group B?

a) 1/9
b) 1/8
c) 1/6
d) 4/17
e) 4/21
CIO
Joined: 09 Mar 2003
Posts: 463
Followers: 2

Kudos [?]: 58 [0], given: 0

### Show Tags

30 Nov 2004, 09:55
i suppose the right answer is A, but actually, the right answer is 5/37, which isn't up there.

This question is based on the known percentages of a normal distribution, and standard deviation. In group A, 16% of the test takers scored below 440, and in group b, 2.5% scored below 440. If we assume 100 students in each class, means 18.5 total scored below 440, so the answer is 2.5/18.5 = 5/37.

In the Princeton Review version of Standard Deviation, they ignore that .5% possibility for convenience sake, so in their world, Class B would have only 2% below 440, making the fraction 2 out of 18, or 1/9.

I don't know why the simplify things like that when it's wrong, but I guess it's a good way to make this stuff taste better to a large portion of their students.

At the same time, I haven't seen a GMAT question that actually required knowing those percentages, and I don't even teach them to my students, so I'm not sure this is even relevant.

Does that sound right?
Director
Joined: 07 Nov 2004
Posts: 689
Followers: 6

Kudos [?]: 142 [0], given: 0

### Show Tags

30 Nov 2004, 10:06
Ian,
Yes, this question is from PR online CAT test and the OA is indeed 1/9.

Manager
Joined: 29 Jul 2004
Posts: 61
Followers: 1

Kudos [?]: 0 [0], given: 0

### Show Tags

30 Nov 2004, 10:17
The problem says the values are normally distributed. Therefore 68% of all values fall within 1 standard deviation, 95% of all values fall within 2 standard deviations, and 99% of all values fall within 3 standard deviations.

Group A has a mean of 460 and a standard deviation of 20. So 1 standard deviation would be 440, so you are looking for values less than 440. Since you are dealing with only the left side of the curve you use 1/2 of 32 or 16%.

Group B has a mean of 520 and a standard deviation of 40. So it would take 2 standard deviations to get to 440 (520-80). Since you are dealing with the left side of the curve you use 1/2 of 5 or 2.5%.

Since you will be taking both 2.5% and 16% of a number use a number both go into like 400. 2.5% of 400=10 and 16% of 400=64. Since you can't have a fraction of a student and you want below scores of 440(not equal to them), use 9 and 63. Now you have 9/(63+9)=9/72=1/8. So the answer is B.

I had to look up the % tables and do a some calculations to get a number that worked well like 400. At first I got 5/37, but then I figured they wanted scores less than 440 so I decreased my values and ended up with an answer that was there. I really hope there isn't a question like this when I take the gmat.
CIO
Joined: 09 Mar 2003
Posts: 463
Followers: 2

Kudos [?]: 58 [0], given: 0

### Show Tags

30 Nov 2004, 10:46
Quote:
Since you will be taking both 2.5% and 16% of a number use a number both go into like 400. 2.5% of 400=10 and 16% of 400=64. Since you can't have a fraction of a student and you want below scores of 440(not equal to them), use 9 and 63. Now you have 9/(63+9)=9/72=1/8. So the answer is B.

Todd,

I agree with everything you wrote except this part. You were right on choosing 400, but you could have gone with 10 and 64, getting the right answer. The choice to drop 1 from each is arbitrary - the standard deviation ranges tell us the percentages BELOW a certain point, so while you can't be sure how many got exactly 440, you can know exactly what percentage got BELOW 440, which is how the question is stated. It could be that 2 or 3 or 4 got 440 exactly, but with a normal distribution we're not talking about exacts, we're talking about statistical probabilities.

Therefore, the answer would have been 10/74, or 5/37.

and to answer the other part of your response, I don't think this will be on the test, but if it is, you seem very equipped to deal with it...
Manager
Joined: 29 Jul 2004
Posts: 61
Followers: 1

Kudos [?]: 0 [0], given: 0

### Show Tags

30 Nov 2004, 13:09
ian7777 wrote:
Quote:
Since you will be taking both 2.5% and 16% of a number use a number both go into like 400. 2.5% of 400=10 and 16% of 400=64. Since you can't have a fraction of a student and you want below scores of 440(not equal to them), use 9 and 63. Now you have 9/(63+9)=9/72=1/8. So the answer is B.

Todd,

I agree with everything you wrote except this part. You were right on choosing 400, but you could have gone with 10 and 64, getting the right answer. The choice to drop 1 from each is arbitrary - the standard deviation ranges tell us the percentages BELOW a certain point, so while you can't be sure how many got exactly 440, you can know exactly what percentage got BELOW 440, which is how the question is stated. It could be that 2 or 3 or 4 got 440 exactly, but with a normal distribution we're not talking about exacts, we're talking about statistical probabilities.

Therefore, the answer would have been 10/74, or 5/37.

and to answer the other part of your response, I don't think this will be on the test, but if it is, you seem very equipped to deal with it...

I guess I was right the first time. I couldn't understand why 5/37 was not an answer choice so I figured I must have understood the problem wrong. That's why I tried to fool with the values. Most standard deviation questions I've run into deal with understanding which group has the larger deviation or something of that nature.
CIO
Joined: 09 Mar 2003
Posts: 463
Followers: 2

Kudos [?]: 58 [0], given: 0

### Show Tags

30 Nov 2004, 22:12
and that's what you can expect on the real test...
Manager
Joined: 30 Jul 2004
Posts: 101
Followers: 1

Kudos [?]: 0 [0], given: 0

### Show Tags

30 Nov 2004, 22:24
good explanations guys ....

was very helpful...to understand SD questions.....
Director
Joined: 21 Sep 2004
Posts: 610
Followers: 1

Kudos [?]: 33 [0], given: 0

### Show Tags

01 Dec 2004, 09:46
toddmartin wrote:
The problem says the values are normally distributed. Therefore 68% of all values fall within 1 standard deviation, 95% of all values fall within 2 standard deviations, and 99% of all values fall within 3 standard deviations.

Group A has a mean of 460 and a standard deviation of 20. So 1 standard deviation would be 440, so you are looking for values less than 440. Since you are dealing with only the left side of the curve you use 1/2 of 32 or 16%.

Group B has a mean of 520 and a standard deviation of 40. So it would take 2 standard deviations to get to 440 (520-80). Since you are dealing with the left side of the curve you use 1/2 of 5 or 2.5%.

Since you will be taking both 2.5% and 16% of a number use a number both go into like 400. 2.5% of 400=10 and 16% of 400=64. Since you can't have a fraction of a student and you want below scores of 440(not equal to them), use 9 and 63. Now you have 9/(63+9)=9/72=1/8. So the answer is B.

I had to look up the % tables and do a some calculations to get a number that worked well like 400. At first I got 5/37, but then I figured they wanted scores less than 440 so I decreased my values and ended up with an answer that was there. I really hope there isn't a question like this when I take the gmat.

can you please explain how u got 32(left side of the curve)
Manager
Joined: 30 Jul 2004
Posts: 101
Followers: 1

Kudos [?]: 0 [0], given: 0

### Show Tags

01 Dec 2004, 11:10
68 % of the data is within 1 Standard Deviation...that means

that values between 460 -20 = 440 and 460 + 20 = 480 will fall for all the 68 % of the data. The rest 100-68 = 32 % falls below 440 and above 480.....since we have to calculate below 440 ..we would take 32/2 = 16 %

(Since it is normal distribution , we can divide 32/2 for values below 440)

hope it helps......
01 Dec 2004, 11:10
Display posts from previous: Sort by