The Washington Post's Valerie Strauss broke the story Sunday that the Democratic Congress is likely to vote to end the National Reporting System, the test that's been given twice a year to a sampling of Head Start students to measure the effectiveness of the program. The story does a good job of explaining a difficult but essential concept of testing--that assessments need to be field-tested to determine if they measure what they're supposed to measure. Concerns have been raised in many quarters about the validity of the National Reporting System. Even the Bush Administration, which pushed for the assessments and which wants the tests to continue, acknowledges the problems.

One of the most vocal critics has been Sam Meisels of the Erikson Institute, and Strauss quotes him in her story. Meisels, who is a member of a national task force on accountability convened by the Pew Charitable Trusts, has twice spoken at Hechinger Institute events about his concerns. He illustrates his criticisms with a description of a test item that he argues is developmentally inappropriate and contains a distinct class bias.

The teacher giving the test shows the child four line drawings--including a wine decanter, a trophy "cup," an ice bucket, and a vase. The testgiver is supposed to ask the kid which of the drawings is a flower "vahz," pronounced to rhyme with "oz." Meisels likes to quip that he's probably used each of the objects as a vase. But he also objects to the preferred pronunciation. Meisels talks about a test item that shows four drawings of faces and the child is asked which one shows "horrified." It's difficult, he says, for a child to distinquish between afraid or angry and "horrified." Sounds like Congress is starting to listen to Meisels and others, who include Edward Zigler.