IMPORTANT DOCUMENTS

Letter to Chancellor Joel Klein
April 29, 2004

Dear Chancellor Klein:

We write to clarify the misinformation you have distributed presumably on the advice of your test advisors. As parents we, too, have consulted test experts across the country.

We believe that your misinformed decision-making and public attacks based on this misinformation are extremely harmful to our children. We want you to do what is best for our children and not become defensive, like many public officials who try to fortify an indefensible position.

Social promotion is not the issue. Basing life decisions on one standardized test, a fallible test, produced by fallible human beings is the issue. We oppose decision-making by tests even with the additional of a last minute cumbersome appeals process.

Because of our concern about using the third grade test as a high stakes instrument we asked for information about the test itself. It wasn't easy to obtain.

Why was this information so difficult to obtain?
We requested the test's Technical Report - a report that all test makers are obligated to provide for public review. Harcourt Measurement told us this document had been "embargoed" by the DOE. The Report was released only after repeated requests by members of the City Council. Why did your Department make it so difficult to obtain? Why did your Department delete passages in the Report? Does the Department have something to hide?

What's in the Report?
A norm-referenced test is one that uses material that a percentage of students will fail. The Technical Report provides evidence that the 3rd grade test is essentially a norm-referenced instrument. (page 25, table 13). Did your test advisors make that clear to you?

The test reflects a bias towards white children.
The test makers report (pg 15, table 5) that field-tested questions favor white test takers. 11 questions favor white children; zero favor Hispanic children; 9 questions favor white children; zero favor African American children. Did your test advisors explain that information?

The Technical Report presents no evidence on how passing scores for the test have been set (pg 23, table 11). Test experts regard such an omission as contrary to professional standards. The passing score of 20 (raw score points) has been set. How was that done? The Report doesn't say. A passing score of 20 would imply that approximately 16,900 of NYC's 80,000 third graders will fail. Did your advisors tell you that?

The Technical Report defines a large error of measurement (pg 23. table 2). The test company reports that the test is accurate to within three questions. If the "passing score" were 17, this would mean that 11,500 would fail. If the 'passing score was raised 3 points, 22,800 would fail. So the error of measurement on the ELA test could lead to a swing of plus or minus 10,000 children failing the test and being "flunked" to repeat grade 3. Did your test advisors alert you to this?

Dr. Walter Haney, a nationally known test expert once hired by Commissioner Mills to review the Regents exams who reviewed the Report states that the Technical Report does not provide "even a scintilla of evidence to support the use of the ELA test as a reasonable basis on which to control promotion of students from grade 3 to grade 4."

Moreover, it says nothing about how the test matches New York State learning standards.

Who supports the use of high stakes tests?
Aside from your Department and some in the State Education Department it turns out that no reputable academic or national educational organization supports the use of high stakes tests -- not the American Education Research Association, not the National Academy of Science, not the National Council of Teachers of English, not even the US Office of Education's Division of Civil Rights endorses high stakes tests -- fallible tests where the results determine critical decisions such as promotion or graduation.

Thus, both the test and the appeals process you have put in place are based on a flawed policy and subject young children to needless suffering and ultimately to unjust failure.

It is time to stop focusing on testing and concentrate on providing our children with quality education.

Sincerely,
Jane Hirschmann

Read more important documents