Chronology Current Month Current Thread Current Date
[Year List] [Month List (current year)] [Date Index] [Thread Index] [Thread Prev] [Thread Next] [Date Prev] [Date Next]

[Phys-L] Logical Thinking and the NEW GRE



New GRE and logical thinking.

Looking at the GRE test, I did not see more than a couple of questions on
proportional reasoning, no questions on conservation reasoning, no questions
on two variable reasoning, no questions on statistical reasoning, no
questions on control of variables, no questions that involve reasoning about
objects that you can not see. Now, I may not have looked hard enough!

It does include some items which involve formal reasoning. But formal
reasoning rises slowly and is probably still quite low at the end of
college. In addition Lawson has proposed that it is not actually necessary
for most needs. There are also some items where asking a "what if" question
is necessary, but many are actually simple math questions posed in a manner
to make them appear difficult.

I would say that this test is missing a large number of types of questions
which reveal the ability of students to think analytically. Actually a few
questions involving what is a dependent vs what is independent variable
might also be useful.

I would also note that they habitually use X and Y as variables in the GRE,
but students have difficulty with using other symbols, so the test needs to
avoid X and Y except where it is necessary such as in the slope example.
The inability to use variables other than X or Y is a big indicator of
concrete thinking. I would note that the math books and teachers play into
this by almost exclusively using X and Y. Then they unsuccessfully try to
tell students to use other variables. Wouldn't it be much easier to
introduce variables first as words, then simplify to single letters and
sometimes 2 letters. Then instead of 2 letter variables introduce the idea
of a subscripted letter. Then X and Y could be introduced for the number
plane, but first introduce it as variables on a map.

Indeed the test draws reading examples from subjects, except for science. I
would call that also very deficient.

A good test for reasoning should not involve too much specialized
vocabulary. The questions about standard deviation and mean may be asking
for specialized math vocabulary that some students may not have had. The
types of questions that were left out can be asked without any very
specialized vocabulary as Lawson did on his test.

I would say that as a true test of reasoning ability it is probably not very
good.

The NYTimes article on it is available for free if you register from:
http://www.nytimes.com/2006/01/08/education/edlife/quiztop.html?ex=113799240
0&en=a4655f4c26b702e1&ei=5070

The complete new GRE is available at:
http://www.ets.org/Media/Tests/GRE/pdf/LargePrintGreFamiliarization.pdf

John M. Clement
Houston, TX
_______________________________________________
Phys-L mailing list
Phys-L@electron.physics.buffalo.edu
https://www.physics.buffalo.edu/mailman/listinfo/phys-l