The Test Generation

In Harrison Distict 2–a low-income, predominantly Latino school district in Colorado Springs–students at every grade level sit for pencil and paper tests in every subject, including art, music, and even physical education. 

On their gym class exam, first-graders are asked to draw a picture of how to position's one's hands while catching a ball. In art, they are asked to identify three shapes within this Picasso cubist portrait. In music, they clap along to the beat of a song as it is played aloud. 

The purpose of these tests is to collect student achievement "growth" data in order to assess teachers and award them merit pay. The typical Harrison child will sit for standardized exams about 25 days each school year.

It seems self-evident, and yet it's rarely discussed: If we are going to evaluate every single teacher, at every grade level, according to how well he or she "grows" student achievement, children are going to have to be assessed far more frequently–and in many more subjects–than they currently are.

To explore the affect of value-added teacher evaluation on kids in the classroom, I spent much of my first semester as a Spencer Fellow reporting on Colorado. The result is my May American Prospect cover story, "The Test Generation." 

In response to Race to the Top, the Colorado state legislature passed one of the most aggressive new teacher evaluation laws in the country, the "Great Teachers and Leaders Bill." Fifty-one percent of every teacher's evaluation will soon be based on student achievement data, and if a teacher fails to grow student achievement two years in a row, he or she can lose tenure protections. 

Local districts will have some autonomy in how they choose to collect this data, but a few dominant models are emerging–and one of the most celebrated by politicians and education reformers is Harrison Distrct 2's, which over-relies, in my view, on testing. I hope you read the entire piece.

One thought on “The Test Generation

  1. Josh R.

    “It seems self-evident, and yet it’s rarely discussed: If we are going to evaluate every single teacher, at every grade level, according to how well he or she “grows” student achievement, children are going to have to be assessed far more frequently–and in many more subjects–than they currently are.”

    A question: value added refers to gains from past performance on the same subject. But is there any research that looks at whether teacher quality in other contemporaneous subjects affects performance?

    So, for instance, let’s say a child is in five classes. Let’s say the math teach is average (whatever that means) but the four other teachers vary consistently, so that they are either all excellent, all average, or all below average. Is there research suggesting that the student in the first condition will perform better on the math test, holding all else constant, than children in the other two conditions?

    Reply

Leave a Reply

Your email address will not be published. Required fields are marked *

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>