| View previous topic :: View next topic |
| Author |
Message |
jamesk486
Really wants to Beat The GMAT!
Joined: 28 Mar 2007
Posts: 137
Thanks given: 0
Thanked 0 times in 0 posts
|
 Posted: Sat May 12, 2007 9:12 pm Post subject: gmat prep question |
 |
|
m>n?
(1) m/n>1
(2) (m-n)/n>(m-n)/m |
|
| Back to top |
|
 |
|
|
Neo2000
Moderator
Joined: 27 Jan 2007
Posts: 363
Thanks given: 16
Thanked 14 times in 14 posts
Location: India
Test Date: 30/09
Target GMAT Score: 710
|
| Posted: Sun May 13, 2007 12:45 am Post subject: |
|
|
Only statement 2 suffices.
Statement 1 does not suffice since we do not know the values of m and n
i.e. m & n can be either positive e.g. 8 and 4
or negative -8 and -4 |
|
| Back to top |
|
|
jamesk486
Really wants to Beat The GMAT!
Joined: 28 Mar 2007
Posts: 137
Thanks given: 0
Thanked 0 times in 0 posts
|
| Posted: Sun May 13, 2007 2:47 am Post subject: |
|
|
| could you explain condition 2 pls? |
|
| Back to top |
|
|
jayhawk2001
Moderator
Joined: 28 Jan 2007
Posts: 789
Thanks given: 0
Thanked 11 times in 11 posts
Location: Silicon valley, California
|
| Posted: Sun May 13, 2007 7:53 am Post subject: |
|
|
| Neo2000 wrote: | Only statement 2 suffices.
Statement 1 does not suffice since we do not know the values of m and n
i.e. m & n can be either positive e.g. 8 and 4
or negative -8 and -4 |
I don't think 2 suffices as well.
Take m=4, n=2. (2) satisfies condition (4-2)/2 > (4-2)/4 and m > n
Take m=-4, n=-2. (2) satisfies condition (-4+2)/-2 > (-4+2)/-4 but m < n
So, I think it should be E |
|
| Back to top |
|
|
atarafder
Just gettin' started!
Joined: 12 May 2007
Posts: 17
Thanks given: 0
Thanked 0 times in 0 posts
|
| Posted: Sun May 13, 2007 10:07 am Post subject: |
|
|
(m-n)/n > (m-n)/m
multiply both sides by mn/(m-n)
you get m> n, so I think 2 should suffice. |
|
| Back to top |
|
|
|
|
atarafder
Just gettin' started!
Joined: 12 May 2007
Posts: 17
Thanks given: 0
Thanked 0 times in 0 posts
|
| Posted: Sun May 13, 2007 10:11 am Post subject: |
|
|
actually I realized I can do the same approach with 1 too -
m/n >1
multiply both sides by n and you get m>n
So should we test with some random values instead of this approach? Any ideas? |
|
| Back to top |
|
|
jayhawk2001
Moderator
Joined: 28 Jan 2007
Posts: 789
Thanks given: 0
Thanked 11 times in 11 posts
Location: Silicon valley, California
|
| Posted: Sun May 13, 2007 11:40 am Post subject: |
|
|
| atarafder wrote: | (m-n)/n > (m-n)/m
multiply both sides by mn/(m-n)
you get m> n, so I think 2 should suffice. |
You can multiply both sides by mn/(m-n) but you don't know what to do
with the inequality sign -- does it reverse in direction or does it remain
the same ?
Put another way
(m-n)/n > (m-n)/m
If m-n is positive, you can do
1/n > 1/m
else, it becomes
1/n < 1/m |
|
| Back to top |
|
|
|
