Since reading the article "NYC schools eye math tests for kindergartners," which appeared in the Watertown Daily Times Aug. 24, I have been wondering whether the author of this experiment is completely ignorant of sound educational practices for young children, or if he just hates 5-year-olds. To subject a kindergartner to any test of 60 to 90 minutes is reprehensible!

I note that the tests will be administered either as workbook-type tests or executed on the computer, both of which will undoubtedly be machine-scored. Such tests only fortify what every early childhood educator knows, which is that tests imposed on 5-year-olds are unreliable.

I recall a child in my kindergarten class being asked to put an "X" on a picture of a cherry, which he did not do. Yet when I asked him to point out the cherry, he did. I asked him why he didn't put an "X" on it, and he replied, "But I like cherries."

My questioning him gave me insight into his perfectly sound reasoning, yet if it had been machine scored, no "X" on the cherry would have been considered a wrong answer. This is a perfect example of the unreliability of two dimensional, machine-scored tests for young children. In spite of what James Liebman claims in the article, these tests will indeed affect pupil grades and pupil placement, and they will create anxiety for the children subjected to such inappropriate demands. I urge the principals of NYC schools to consult with early childhood educators before considering using these tests.

Mary Alice B. Robinson

Watertown