begin  quoting Andrew Lentvorski as of Sun, Mar 18, 2007 at 11:15:44PM -0700:
> Stewart Stremler wrote:
> 
> >Hm. Then I wonder what the appeal was based on.
> 
> "If you have two triangular pyramids (four sides each) and join them 
> such that one side of one is in full contact with one side of the other, 
> how many faces are left exposed?"
> 
> Four sides each can be read as either "four sides other than the base" 
> or "four sides including the base".  If "four sides including the base", 
> then sticking two faces together gives 6 exposed faces.  If "four sides 
> other than the base" then sticking two identical faces together gives 8 
> exposed faces.

Ah, okay.  A side is not necessarily a face.

I really don't like the term "triangular pyramid" either. :)

> However, unless somebody can quote *verbatim* the wording of the 
> question, it really isn't worth considering as there are simply too many 
> permutations.

Yup.

> This is one of the things which the SAT really gets correct with their 
> "experimental" sections.  They clean out the questions which are 
> misleading to the better students.
> 
> I bumped into this a *lot* when I was preparing for the SAT.  Whenever I 
>  took a sample test from J. Random Prep Book, I could never score more 
> than about 1400.  Never.  When I took the SAT, I rather easily scored 1560.

Oddly enough, I never took the SAT.  The ACT was sufficient the first
time I went to school (for the schools I was looking at), and the second
time I counted as a transfer student...

> The difference was that the SAT questions are *never* misleading.  Hard 
> questions are hard.  Easy questions are easy.  Everybody gets the easy 
> questions.  Only good students get the hard questions.  Any hard 
> question which easy students might get because of some random change 
> correlation get thrown out after statistical testing identifies that in 
> the question.

Writing hard-but-not-misleading questions is very, very hard.

-- 
Some people are better at it than others.
Stewart Stremler

-- 
[email protected]
http://www.kernel-panic.org/cgi-bin/mailman/listinfo/kplug-lpsg

Reply via email to