Read Dr. Stephen Krashen's response to an article by Walter Isaacson
titled, "How to Raise the Standard in America's Schools," appearing in Time Magazine on Wednesday, Apr. 15, 2009. I agree with Dr. Krashen that we can find out what works in education better through other means and also that we don't have to conduct census testing for us to find out what works.
Do we need national standards?
Sent to Time Magazine, April 24, 2009
Re: "How to raise the standard in America's schools," Time, April 27,
Walter Isaacson argues that we need uniform national standards so we
can find out what works in education and compare our students to those
elsewhere [April 27]. Neither argument holds.
Progress in educational research does not require uniform tests given
to every student in the country. Just as research is done in many
other areas, educational researchers compare small groups of students
who differ only in the treatment under examination, using highly
reliable and valid measures, and then generalize from these results.
This approach has been very successful, and is a much better method
than testing millions of children, trying to deal with all the factors
other than the treatment that could affect results.
Similarly, if the purpose of nationally standardized tests is to see
how well our children are doing, we do not need to test every child.
We need only a representative sample. When your doctor takes your
blood, he takes only a small amount. He doesn't need all of it.
Stephen Krashen, Ph.D.
University of Southern California