willget800
rhyme.. I feel bad for you because that was the last topic discussed in OG on Pg 139 of the book, and they give only one example of it in the entire book. No OG problem in the book discusses this.. but if you want to get to pg 139 of the OG.. it says potentially how to solve these questions..
in this case I am assuming, we would try to solve like this...
y = f(x) = 3/4 - 2x^2
find y intercept, x intercept..
so y = 0, x = +or-sqrt(3/8)
when x = 0, f(0) = 3/4.
I guess then just plot these three points atleast and a few more points... and then draw something up..
does this help? I would be stunned to see this question as the first question of my exam.. do u think that put u back? or was it more than that?
anyway thanks a lot for sharing this information.. its an eye opener for the rest of us to atleast do a few problems on quadratic eqns which generate parabolic graphs!
all the best.. i am sure you will be more prepared next time for these out of context questions!!!!!!!!!!!
It was only necessary to determine whether the y intercept was positive (that eliminated 2 options), then determine whether the coefficient on x was positive or negative, (its negative), in which case that eliminates 1 more option, leaving you with a curve that has a y intercept above zero, is convex and either has x intercepts equidistant from zero or not.
It's was NOT necessary to actually find the exact values of the x intercepts (or for that matter the exact value of the y intercept). It was sufficient to determine their sign and relative positions. In fact, I know that the y intercept was something like 1/3, and the x intercepts were something like 1 and 1/4 and -1 1/4 ... not something you can easily identify on a small graph.
Absolutely, it threw me. I wasn't scared by it - that is, I didnt feel like "OH MAH GAWD, WHAT THE !@#! IS THIS?" when I saw it, but I did feel like "Hmmm.. this is a kinda new... I guess I'll pick points..."
Theres no question that, as a first question, it was not a confidence builder.
Drummond is also right that "what I knew" vs "What I was feeling" is a good point - and I agree that much of what looks easy.... etc... but when you get a question asking you what 10% of .0002 is, there's little doubt that this is not going to be a question that the vast majority of people get wrong.
Obviously, every question has its own bell curve, but after having done a bunch of percentage q's, you begin to identify which ones are clearly more difficult (and therefore logically have a bell curve indicative of this) than those where are clearly less complex.
The most basic (10% of 80 is what?)
The slightly less basic (10% of some fraction)
The slightly more complex sucessive percents q (If a price of a product is increased by 10% then decreased by 10% what is the new price relative to the old one?)
The complexity part-whole questions (If the number of people increased by 28 from 1997 to 1998, and there were 209 people in 1998, by what percent did the population increase from 1997 to 1998?)
The complex multiple-subject questions (If two containers are empty and x liters of water is poured into both containers making one container 1/4 full and one 1/3 full and then the smaller of the two is poured into the larger of the two, what percentage full is the larger container?)
While I agree that what seems "easy" and "hard" is subjective and not an indicator of how you are doing, there's little doubt in my mind that when you see a question like the one I described, it's pretty obvious its not a hard probability.
All the best. 
