Friday, May 14, 2010

DESELECTION, WAY TO IMPROVE STUDENT ACHIEVEMENT

In 1983, we were warned that we were “A Nation at Risk. Our once unchallenged preeminence in commerce, industry, science, and technological innovation is being overtaken by competitors throughout the world.”

When the Governors met in 1989, they resolved that they would combat the “rising tide of mediocrity” by our being first internationally in math and science by 2000. Many ways to improve student learning have been tried: smaller class size, smaller pupil-teacher ratios, increased teacher experience, increased teacher graduate degrees, and programs galore. Nothing has worked. “What was unimaginable a generation ago, has begun to occur—others are matching and surpassing our educational attainments,” the Governors were warned.

Since the Governors’ 1989 conference, the focus has been on improving teacher effectiveness as the “only viable way to accomplish the governors’ goals.

But how? Teacher hiring and retraining have not worked, in that they have not improved student achievement. “Some teachers do a very poor job, and few people believe that the worst teachers can be transformed into good teachers,” observes economist Eric A. Hanushek, Senior Fellow at the Hoover Institution at Stanford University, who has undertaken to find a solution to this problem. An alternative approach, he suggests, is to remove or “deselect” inefficient teachers.

“How much progress in student achievement could be accomplished by instituting a program for removing, or deselecting, the least-effective teachers?” Hanushek asks.

There is now considerable interest in tying teachers’ salaries to student achievement. Indeed Hanushek, who specializes in the economics of education, goes further. He finds a mathematical relationship between effective teachers and student achievement. It is the subject of his essay, “Teacher Deselection.”

The answer is half-a-standard-deviation (0.5 sd), according to his calculation. (A standard deviation is a measure of the spread of the data from its mean or center. One standard deviation on either side of the mean, above it and below it, includes 68 percent of the cases in a normal distribution of data.)

How bad are American students? HongKong, at the top of the 40-country ranking of Program for International Student Achievement (PISA), is 2/3 of a standard deviation ahead of the average American student. An improvement of half-a-standard-deviation (0.5 sd), calculates Professor Hanushek, would put American students close to the top, on a level with Canadian students who in 2003 were just behind the Netherlands and Japan.

And an increase of half-a-standard-deviation would dramatically grow the U.S. economy, helping to regain our preeminence in commerce, industry, science, and technology, which are closely associated with student achievement.

In analyzing teacher quality, Hanushek finds that “differences in annual achievement-growth between an average and a good teacher in math are at least 0.11 sd of student achievement. Actually, he concludes, that’s conservative. If a student had a good teacher for four or five years in a row, “the increased learning would be sufficient to close entirely the average gap between a typical low-income student . . . and the average student.”

If the average learning-growth is one grade level, the least effective five percent of teachers achieves 2/3 of a grade-level annually. The bottom one percent of teachers achieves ½ of a grade level annually, which, measured from the center or mean, is the 84th percentile.

In a school of 30 teachers, “eliminating the least effective six to ten percent of the teachers would bring student achievement up by 0.5 sd.” Eliminating the bottom two or three teachers would bring the average up to Canada’s students, who are high up on the PISA scale. Several evaluations of high-performing school systems around the world find that “the best schools do not allow ineffective teachers to remain in the classroom for long,” observes Hanushek.

What are the implications for the U.S. economy, asks Hanushek. “Had these policies been in effect in 1989 at the time of the governors’ conference, we would be reaping gains in the national economy; and the gains to Gross Domestic Product would be enough to cover all our spending on K-12 schooling,” he answers.

Presently the Colorado state legislature has passed a bill to tie teacher-tenure in to student achievement, but without being aware of the improvement in student achievement and economic growth which are achievable should the tenure bill pass. The bill provides that new teachers can only be retained if their students have achieved three years in a row. Tenured teachers could lose job-protection if they are deemed “ineffective” in two straight years. Supported by one teachers’ union and opposed by the other, the bill has passed the Senate and the House on Wednesday, and is ready for the Governor’s signature.

Unfortunately escaping the attention of the Colorado Legislature is Hanushek’s 18-paged essay, “Teacher Deselection,” which is Chapter 8 in Creating a New Teaching Profession, edited by Dan Goldhaber and Jane Hannaway (Washington, D.C., The Urban Institute Press, 2009).


By Natalie Sirkin
C2010
Post a Comment