I see that others have now made most of the points I wanted to (AOL is
eating my attempts to send outgoing e-mails from that account), but I'll
summarize what I see as the most relevant:
First and foremost, THE SAT DOES NOT TEST SCIENCE KNOWLEDGE. It has a
Verbal part (vocabulary and reading comprehension) and a Math part. No
science at all. So SAT scores are no measure of scientific knowledge and
the analysis Allen Roy gave is meaningless.
It *might* be interesting to see such a comparison with a widely given test
of science knowledge, if there is one. Even then, there would be major
problems:
1) The Kansas creationist standards were only recently instituted, so have
had no chance to affect anything. One might look for lower science test
scores in 2006 or so if they remain in place. For an effect in the
present, one might look at Alabama, which has had an anti-evolution
"disclaimer" in place for a few years.
2) Test scores are strongly correlated with other factors; in particular a
state's results would depend on the fraction of students from impoverished
or otherwise disadvantaged backgrounds and also the fraction who did not
have English as a first language. On both of these counts, Kansas is
better off than most states. One could examine the same state before and
after a change in educational standards, but even then it would be
difficult to separate that from other factors.
-------------------------------------------------------------------------
| Dr. Allan H. Harvey | aharvey@boulder.nist.gov |
| Physical and Chemical Properties Division | "Don't blame the |
| National Institute of Standards & Technology | government for what I |
| 325 Broadway, Boulder, CO 80305 | say, or vice versa." |
-------------------------------------------------------------------------
This archive was generated by hypermail 2b29 : Fri Oct 06 2000 - 14:49:14 EDT