TehJay
I said (C). I don't understand how it can be (E) - you don't really have the information to determine how many non-subscribers placed orders, which is what (E) is talking about. For example, what if 100 subscribers placed orders, and 30 of them were under age 35, while only 5 non-subscribers placed orders, and 4 of them were under age 35? Then both findings are still correct, but (E) is false (unless we're going on some very arbitrary definition of "many").
No, you may have misinterpreted one of the two findings. The second finding says that most orders were placed by people under 35; that includes orders from
both subscribers and non-subscribers. I think you are interpreting that finding to be about orders from non-subscribers only, but it's not. So in your hypothetical example, you have 105 orders in total, but still, only 34 orders come from people under the age of thirty-five. That isn't consistent with the second finding in the stem which tells us that most orders come from under-thirty-fives, so is not a possible scenario.
TehJay
(C) fits both findings the best. 70% of subscribers who placed orders were age 35 or over, a big majority. The only way to make (C) not true is if the number of non-subscribers who placed orders is larger than the number of subscribers who placed orders - and even then, you can't really determine whether or not the 30% of subscribers under 35 + the "most" of non-subscribers under 35 outnumber the 70% of subscribers + the remaining non-subscribers.
It's actually mathematically impossible for C to be true. Say you have S subscribers and N non-subscribers who placed orders. We know that 0.7S orders came from subscribers over thirty-five. Say you have X orders in total from non-subscribers over 35. We know from the second finding that less than half of all orders come from people over thirty-five, so:
(0.7S + X)/(N+S) < 1/2
But the proportion of all orders coming from subscribers over thirty-five is 0.7S/(N+S), and this is clearly less than the left side of the inequality above, so must be less than one half. So it's impossible for 'most' orders to have come from subscribers over thirty-five, and C cannot be true.
But 'E' may not be true always.
For example let us consider the following figures.
Total orders by non subscribers = 55 ( 50 by <35 age ; 5 by >35 age group) (since nothing was mentioned about the distribution of orders by non subscribers)
Please guide where I made a mistake in assuming the above figures.