|
Author |
Message |
|
TAGS:
|
|
|
SVP
Joined: 03 Jan 2005
Posts: 2322
Followers: 9
Kudos [?]:
157
[0], given: 0
|
The following table shows scores of Class A and Class B in a [#permalink]
 27 Jan 2005, 23:31
Question Stats:
0% (00:00) correct
 0% (00:00) wrong based on 0 sessions
The following table shows scores of Class A and Class B in a test:
A median 80 average 82
B median 78 average 74
Is the average score for all students greater than the median score of all students?
(I)A has 37 students, B has 40 students
(II)There are 77 students in total
|
|
|
|
|
|
|
Director
Joined: 19 Nov 2004
Posts: 589
Location: SF Bay Area, USA
Followers: 2
Kudos [?]:
7
[0], given: 0
|
I think median of a combined set cannot be less than median of each.
If this right, I) would suffice. The information provided in A is required to calculate the average (which will be around 77) and since the median is at least 78, we know the answer.
II) We can't decide on average based on this info, as the average could come close to B's average if there are lost of students in B and less in A.
I quickly pick A.
|
|
|
|
|
|
SVP
Joined: 18 Nov 2004
Posts: 1544
Followers: 2
Kudos [?]:
9
[0], given: 0
|
"A" for me. Not a quick one though. 
HongHu, can u share the source of this problem, may be I will buy that book
|
|
|
|
|
|
Senior Manager
Joined: 25 Oct 2004
Posts: 267
Followers: 1
Kudos [?]:
3
[0], given: 0
|
Had to struggle with this one is it A tho......
|
|
|
|
|
|
Director
Joined: 07 Jun 2004
Posts: 638
Location: PA
Followers: 1
Kudos [?]:
60
[0], given: 22
|
from I we can get the weighted average
A is good enough
|
|
|
|
|
|
SVP
Joined: 03 Jan 2005
Posts: 2322
Followers: 9
Kudos [?]:
157
[0], given: 0
|
banerjeea_98 wrote: "A" for me. Not a quick one though. HongHu, can u share the source of this problem, may be I will buy that book It's a Chinese one. 
|
|
|
|
|
|
SVP
Joined: 18 Nov 2004
Posts: 1544
Followers: 2
Kudos [?]:
9
[0], given: 0
|
HongHu wrote: banerjeea_98 wrote: "A" for me. Not a quick one though. HongHu, can u share the source of this problem, may be I will buy that book It's a Chinese one. I can read chinese  just kidding....
|
|
|
|
|
|
SVP
Joined: 03 Jan 2005
Posts: 2322
Followers: 9
Kudos [?]:
157
[0], given: 0
|
I'm taking my test on 2/15. I can mail it to you after that. :D
(Chinese is very easy.  ;))
|
|
|
|
|
|
Senior Manager
Joined: 25 Oct 2004
Posts: 267
Followers: 1
Kudos [?]:
3
[0], given: 0
|
Hong Hu you seem to have a good collection of questions. Sad its in Chinese
|
|
|
|
|
|
SVP
Joined: 25 Nov 2004
Posts: 1581
Followers: 4
Kudos [?]:
13
[0], given: 0
|
b/w A and E, probably A is the answer.-same reason as did by nocilis....
|
|
|
|
|
|
SVP
Joined: 03 Jan 2005
Posts: 2322
Followers: 9
Kudos [?]:
157
[0], given: 0
|
Oops, did I forget to post OA? It is (A).
|
|
|
|
|
|
Intern
Joined: 26 Jan 2005
Posts: 12
Followers: 0
Kudos [?]:
0
[0], given: 0
|
[quote="nocilis"]I think median of a combined set cannot be less than median of each.
I dont think this statement is correct.
Can anyone explain how is it A?
|
|
|
|
|
|
SVP
Joined: 03 Jan 2005
Posts: 2322
Followers: 9
Kudos [?]:
157
[0], given: 0
|
The median of a combined set can't be smaller than the smallest median of the individual sets, and can't be larger than the largest median of the individual sets.
|
|
|
|
|
|
VP
Joined: 13 Jun 2004
Posts: 1135
Location: London, UK
Schools: Tuck'08
Followers: 5
Kudos [?]:
14
[0], given: 0
|
swath20 wrote: Hong Hu you seem to have a good collection of questions. Sad its in Chinese
Not so sad  Hong Hu are you living in China ? Where did you buy this book ? I am a french guy lost in Shenzhen  but I can speak chinese  so i am interested...
|
|
|
|
|
|
SVP
Joined: 03 Jan 2005
Posts: 2322
Followers: 9
Kudos [?]:
157
[0], given: 0
|
No actually I'm now in US, I bought the book from a Chinese friend that came from Beijing about a year ago, I think. I'm sure you can find GMAT preparation books in China if you ask the right people.
|
|
|
|
|
|
Director
Joined: 29 Oct 2004
Posts: 894
Followers: 2
Kudos [?]:
20
[0], given: 0
|
HongHu wrote: The median of a combined set can't be smaller than the smallest median of the individual sets, and can't be larger than the largest median of the individual sets.
OK, this is the first time I hear this. Can you prove it?
|
|
|
|
|
|
SVP
Joined: 03 Jan 2005
Posts: 2322
Followers: 9
Kudos [?]:
157
[0], given: 0
|
Give me some time to think about how to prove it.
|
|
|
|
|
|
Director
Joined: 29 Oct 2004
Posts: 894
Followers: 2
Kudos [?]:
20
[0], given: 0
|
HongHu wrote: Give me some time to think about how to prove it.
OK, waiting...
Where can I learn more about mean, median, mode, S.D? I just want to know all the properties/theories behind. My math knowledge is not from American education system. Back then I didn't get to know about those and probability until I moved to the US. So there is my weakness.
|
|
|
|
|
|
VP
Joined: 13 Jun 2004
Posts: 1135
Location: London, UK
Schools: Tuck'08
Followers: 5
Kudos [?]:
14
[0], given: 0
|
HongHu wrote: Give me some time to think about how to prove it.
I don't have any algebric formula to explain this but I definitely agree with HongHu. The global median has to be inside the set of the 2 others median, I put some drawing attached with this post, it doesn't explain anything but it just help me to figure it out...
Attachments
median.JPG [ 19.09 KiB | Viewed 187 times ]
|
|
|
|
|
|
Director
Joined: 29 Oct 2004
Posts: 894
Followers: 2
Kudos [?]:
20
[0], given: 0
|
Here is the proof.
Let U(A)=the smallest x so that at least half the values are <x, and
V(A)=the largest x so that at least half the values are >x, then that
makes M(A)=(U(A)+V(A))/2: i.e. no need to distinguish between even or
odd number of elements.
We can assume without loss of generality that U(A)<=U(B): otherwise we
just swap A and B. This gives three different possibilities:
a) U(A)<=V(A)<U(B)<=V(B)
b) U(A)<=U(B)<=V(A)<=V(B)
c) U(A)<U(B)<=V(B)<V(A)
In case a, for x<V(A), no more than half the values of A are <x; since
x<U(B), less than half the values are <x. Hence, less than half the
values of A+B are <x, and we must have U(A+B)>x. As this holds for all
x<V(A), this makes V(A+B)>=V(A). Similarly, U(A+B)<=U(B). Hence,
M(A)<=V(A)<=M(A+B)<=U(B)<=M(B).
If U(A)<=U(B)<=V(A), for x<U(B) less than half the values in B are <x,
whereas x<V(A) makes at most half the values of A <x; thus less than
half the values of A+B are <x, which makes x<U(A+B). However, for
x=U(B), at least half than values of both A and B are <x, so this is
also true for A+B. This proves that U(A+B)=U(B).
Similarly, if U(B)<=V(A)<=V(B), then V(A+B)=V(A). Case b now follows
as 2*M(A)=U(A)+V(A)<=U(B)+V(A)=U(A+B)+V(A+B)=2*M(A+B) etc.
In case c, the above arguments gives U(A+B)=U(B) and V(A+B)=V(B),
which makes M(A+B)=M(B).
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
close
