|
|
Re: Engage To Excel: Producing One Million Additional College
Graduates with STEM Degree
Posted:
Apr 13, 2012 2:57 PM
|
|
On Fri, Apr 13, 2012 at 1:21 PM, jk@israeliteknight.com
<jk@israeliteknight.com> wrote:
> Your statements were already strange enough, but now we have this:
>
>
> > "It is not an odd statement - it is the truth. The powers-that-be in the US in 1995 made the wrong decision to send to the TIMSS Advanced test students that never took the courses whose subject matter was being tested. The vast majority of those US students that were sent to take the physics test never took a physics course. That is unfair to these students - they at the very least should have sent only those students that took a physics course. The same for calculus - the vast majority of those US students that were sent to take the test of which 25% of the questions covered calculus topics never took a calculus course. That is unfair to these students - they at the very least should have sent only those students that took a calculus course."
> >
>
> You don't seem to understand how TIMSS is SUPPOSED to work. It is DESIGNED as a random sampling of the student population to enable comparisons between different countries' education systems, which means the education mafia CANNOT just "send to the TIMSS Advanced test students that never took the courses whose subject matter was being tested".
>
(To the reader: My post
http://mathforum.org/kb/message.jspa?messageID=7765365
is my message to which the above is a reply.)
It does not mean that.
Taking a random sample of the top 14% of one country (the US) to
compare to a random sample of the top 2% of another country (the
Russian Federation) breaks every rule with respect to fair
comparisons, especially when only the top 2% of each of these two
countries studied that which was being tested. (For the US, the vast
majority of that 14% never took calculus even though 25% of the test
questions were calculus questions.)
What if the situation were reversed in 1995 and it was the Russian
Federation that sent a random sample of their top 14% and the US sent
a random sample from its top 2%?
What would have happened is this:
Since we know that according to the Russian Federation itself that
only its top 2% ever took even just pre-calculus, we know that it
would have been an absolute disaster, much worse than even the US
where at least most of that 14% took at least pre-calculus.
And we know that the top 2% of the US at that time would have scored
higher than even the average of the highest scoring country in 1995
(and in 2008 as well). We know this because, again, we have this
scientific study giving the TIMSS Advanced math test to those who took
an AP Calculus test (not just those who took an AP Calculus course),
where these students in 1995 were roughly our top 2% (in 1995, out of
roughly maybe 4 million people of that age at that time including
dropouts, we had not far from 100,000 taking an AP Calculus test):
"How Well Do Advanced Placement Students Perform on the TIMSS Advanced
Mathematics and Physics Tests?"
http://apcentral.collegeboard.com/apc/public/repository/ap01.pdf.ti_7958.pdf
This shows via a re-administering of TIMSS Advanced to only AP
Calculus students that took an AP Calculus exam that if the US sent
only those students that were in AP Calculus and took an AP Calculus
exam, then we can expect that the US would have scored highest in the
world among all those countries that took the test in 1995 but also
that took it in 2008 (the US like most other countries from 1995 did
not take the test in 2008). If the US sent only those who passed an AP
Calculus exam, then we can expect that the US would have scored much
higher than the rest of the world that took the test in either year.
Here is the proof:
Exhibit 5: Average Achievement of AP Calculus Students in Advanced
Mathematics by Results AP Calculus Examinations:
Less than 3 on AP Calculus AB 565 (TIMSS Advanced math average score)
3 or better on AP Calculus AB 586 (TIMSS Advanced math average score)
Less than 3 on AP Calculus BC 564 (TIMSS Advanced math average score)
3 or better on AP Calculus BC 633 (TIMSS Advanced math average score)
The average of all of them put together was 573 and the average of
those who passed with a 3, 4, or 5 was a 596.
France had the highest score in 1995 with a 557 and the Russian
Federation had the highest score in 2008 with a 561 (the latter scored
near the top close to but not quite as high as 557 in 1995).
On top of all that, those who take an AP Calculus exam (they scored
573 in that study) are not now roughly the top 2% but are now roughly
the top 7.5%, and on top of that, those who pass their AP exam (they
scored 596 in that study) are not now roughly the top 1.4% but are now
roughly the top 4.6%. Compare all that to the fact that although the
Russian Federation sent only their top 2% in 1995, they sent only
roughly their top 1.5% in 2008, when they scored that 561, which is
less than what our now 4.6% and 7.5% would be expected to score, which
is respectively 596 and 573. Our top 1.5% would score even much higher
than our 4.6% at 596, meaning that now, our top 1.5% would blow away
the top 1.5% of the Russian Federation.
(The above study also shows that if the US sent only those who took a
physics AP exam, then the US would have scored in second place.)
|
|
