Here

let's take the numbers from 1 to 5

So the average of the numbers = \(\frac{15}{5} =3\)

Now if we remove a number what will be the average? to get that we need to take the greatest number from the series

So in this case , 5 is the greatest

So average of 1 ,2 ,3 ,4 =\(\frac{10}{4}= 2.5\)

Therefore the percentage decrease = \(\frac{{3 - 2.5}}{3} * 100 = 16.66%\), this the maximum value we can get if we remove from the set.

Hence QTY A < QTY B

*** it can be any 5 consecutive positive numbers the maximum value will be 16.66%****
_________________

tricky wording. Only the decreases are issues .

How do we know which number is to be excluded ? Ans can change depending on the number we exclude

@GreenlightTestPrep

Here is an algebraic solution that allows us to examine all possible cases.

Let x = the smallest integer in the set
So, x + 1 = the next consecutive integer
x + 2 = the next integer
x + 3 = the next integer
x + 4 = the greatest integer

Average \(= \frac{x+(x+1)+(x+2)+(x+3)+(x+4)}{5}=\frac{5x+10}{5}=x+2\)

Key concept: In order to get the greatest DECREASE in the average, we must remove the biggest number in the set.
So we'll remove (x+4) from the set
The new average \(= \frac{x+(x+1)+(x+2)+(x+3)}{4}=\frac{4x+6}{4}=x+1.5\)

Percent decrease = (100)(old - new)/old

We get: percent decrease \(= \frac{(100)[(x+2)-(x+1.5)]}{x+2}\)

\(= \frac{(100)[0.5]}{x+2}\)

\(= \frac{50}{x+2}\)

Notice that the percent increase depends on the value of x.
For example, if \(x=8\), then the percent decrease \(= \frac{50}{8+2}= \frac{50}{10}=5%\), which is LESS THAN 20%

Also notice that, in order to maximize the percent decrease, we must minimize the value of x.
Since we're told x is a positive integer, the smallest possible value of x is 1.
When \(x=1\), then the percent decrease \(= \frac{50}{1+2}= \frac{50}{3}≈16.666...%\)
This means 16.666...% is the GREATEST possible value of Quantity A.

This means Quantity B will always be greater than Quantity A

Answer: B

Cheers,
Brent
_________________

Great question!
Notice that, if we remove the middle number, then the percent decrease in the average is zero.
So, in this case, Quantity B will be greater.
At this point our goal should be to MAXIMIZE the percent decrease. This is achieved by removing the biggest number in the set.
As I show in my post above, removing the biggest number in the set will always yield a percent decrease that is less than 20%.

Cheers,
Brent
_________________

Let the 5 numbers be: \(x, x+1, x+2, x+3, x+4\)

Since the numbers are in Arithmetic Progression (i.e. consecutive numbers have a constant gap), we have:
Mean = Median = 3rd term = \(x+2\)

(Note: You can always add up the terms and check the average)

One of the 5 numbers will be dropped

Maximum change in mean will occur if either of the 2 extreme terms, i.e. \(x\) or \(x+4\) is dropped. The average will decrease if \(x+4\) is dropped (Note: if \(x\) is dropped, the average would actually increase. Also, if the middle number, i.e. \(x+2\) is dropped, there will be no change in the mean)

If \(x+4\) is dropped: The 4 terms are: \(x, x+1, x+2, x+3\)

=> New Mean = Median = \([(x+1)+(x+2)]/2 = x+1.5\)

=> Percent decrease in mean = \([{(x+2)-(x+1.5)}/(x+2)] * 100\) = \([50/(x+2)]%\)

The above percent will be maximum if the value of \(x\) is minimum, i.e. \(x=1\)

=> Maximum percent decrease = \([50/(1+2)]% = 16.67%\)

Thus, Quantity B is greater than Quantity A

Answer B


Note: Some important results that come up here:

In an Arithmetic Progression i.e. consecutive terms having a constant difference of \(d\):

#1. Mean = Median
#2. The average remains unchanged if the middle term (or both middle terms) are removed
#3. The maximum change in mean occurs when one of the extreme terms is removed

#4. The maximum change in mean (when one of the extreme terms is removed) = \(d/2\), where \(d\) is the constant difference between consecutive terms
_________________

I don't understand why we have to decrease; I do not read the question in a way that states that as a rule. Is it possible to see a positive increase as a negative decrease? Would the GRE risk this issue? The entire micro-intention of the test is separate the people who know from the people who don't know and playing word games can produce a knower with a wrong answer and an Edwin J. Goodwin with the right one which doesn't serve GRE's purpose (or those of the admissions offices)

Hello from the GRE Prep Club BumpBot!

Thanks to another GRE Prep Club member, I have just discovered this valuable topic, yet it had no discussion for over a year. I am now bumping it up - doing my job. I think you may find it valuable (esp those replies with Kudos).

Want to see all other topics I dig out? Follow me (click follow button on profile). You will receive a summary of all topics I bump in your profile area as well as via email.