The s subjects in an experiment are divided into 4 groups: 3

[ShahidB174]

Source: GmatHacks

The s subjects in an experiment are divided into 4 groups: 3
test groups and a control. If each group is further divided into
units consisting of u subjects each, with each unit assigned to
a different researcher, how many researchers are assigned to
units?

(A) s/u

(B) u/s

(C) 3s/u

(D) 4s/u

(E) 4u/s

In the above problem when we pick numbers say S=32, then as given we have 4 groups, leading to 8 members per group. Now when we choose U=2 we get 12 (2 x 2 x 2 First Group 2 x 2 x 2 Second Group, 2 x 2 x 2 Third Group and 2 x 2 x 2 Fourth Group) researchers. whereas when we pick U= 8 we get 4 researchers and S/U as answer. Why do we get different answers with the same approach.

[RonPurewal]

if you divide 8 people into groups of 2, you get four groups (not three).

[RonPurewal]

also note that the division of the 's' people into 4 groups is completely irrelevant here, since you're eventually dividing everyone into groups of 'u' people anyway. so, you can shorten the whole process by just dividing 's' people into groups of 'u' people each.

in this sense (= the inclusion of irrelevant information), this problem is rather unrepresentative of the GMAT; it's VERY rare for a multiple-choice GMAT problem to contain any information that isn't actually necessary.