brain sizes: Einstein's and women's

Bob LeChevalier lojbab at lojban.org
Thu Aug 1 10:05:59 EST 2002


"Parse Tree" <parsetree at hotmail.com> wrote:
>> For the 12th grade math questions, N for the US was 2795.  Only 2
>> other countries had a sample size over 2000, and Cyprus had the
>> smallest sample at N=190.
>
>Ha ha ha!!  That's not really big enough to get an accurate picture.
>
>I assumed that it was simply given to every student in the country that was
>of the required grade.

No, TIMSS used sophisticated sampling techniques, and only a couple of
countries managed to meet the sampling criteria, especially for the
12th grade test, which is one reason why the TIMSS-R tested only lower
grades.

The 12th grade test was "final year of schooling", and thus was
different age kids in different countries.  It was a test primarily in
calculus and physics, which in most of this country are electives
taken by only a small percentage of kids (somewhat under 10% take
calculus as I recall), and indeed in order to meet other sampling
criteria, the US had to include some precalculus kids in the test
sample (and those kids are pretty assured to miss the calculus
questions).

The organization conducting the test had no means to require anyone to
participate and indeed, on most national and international testing,
entire states do not participate (NAEP, which is the national
benchmark testing in several subjects, is skipped by a third of the
states, a different third depending on the subject and the grade
level).

This in fact is one reason that the US is NOT as well off in
international competitions as it might be - we have NO national
education policy.  The states are entirely free to do whatever they
want educationally, and the only clout that the Feds have on policy is
a carrot and stick one based on giving big grants with lots of strings
attached.  

lojbab



More information about the Neur-sci mailing list