Teaching used to be easier than it is now. Teachers presented the information, assigned homework, made up tests, and graded students. It was the teacher's job to separate the wheat from the chaff, and the test was one of the most effective tools for that important task. Tests were sometimes complicated or tricky. Students who were weak readers found that their reading difficulties depressed their grades in all subjects, because only the best readers could negotiate the complicated test formats often used in social studies and science classes. However, it was an easy way to grade. Tests were constructed to have ten items, not nine or eleven; or twenty, not nineteen or twenty-one. This made grading easier: "-2" to "-0" was an A, because those grades were 90% or higher.

The way teachers graded affected the way they wrote their tests: that 20th question might not have been important; that 21st question might have been...

If the goal of testing is to find out what a student has learned, then using the older styles of grading doesn't pass the grade itself. Howard Gardner's view of multiple intelligences gives teachers a clearer picture of just how much intelligent students vary.
The new ways of evaluating student progress is complicated, and teaching is already a complicated job. For projects, teachers can use checklists, matrices or rubrics to grade, and this will also assure that students know exactly what is expected. It places some decision-making about what grade will be earned right in front of the student, who can choose to slack off on a project, but will know exactly what it will do to his or her grade.

Using checklists, matrices or rubrics also actually simplifies grading of…