Published Online:

3-Year Effort To Create a 'New Generation' Of Teacher Assessments Nears Completion

Article Tools
  • PrintPrinter-Friendly
  • EmailEmail Article
  • ReprintReprints
  • CommentsComments


By Ann Bradley

Lee S. Shulman, professor of education at Stanford University, has completed his three-year research into how to evaluate the essence of exemplary classroom teaching.

The aim of the Teacher Assessment Project at Stanford has been, he said, to create prototypes for "a new generation" of assessments that will mirror the complexity of teaching and contribute to improvements in teaching and teacher education.

The endeavor was sponsored by the Carnegie Corporation of New York, and has been watched closely by members of the National Board for Professional Teaching Standards.

Public officials have warned that the board's plan to offer voluntary national certification to skilled teachers would depend heavily on its being able to create credible new methods of assessing teacher performance.

Although the national board is also funded by Carnegie, Mr. Shulman's work was not done for the board. He describes the two projects as "parallel developments that, while not linked formally, are certainly linked philosophically."

The Stanford researchers, who began their project in June 1986, first developed sample exercises that could be used over two days in an "assessment center" for teachers.

The 19 exercises focused on teaching equivalent fractions to elementary-school students and the American Revolution to high-school students.

The exercises were field tested by 40 teacher candidates in the summer of 1987. Then the researchers worked to develop a scoring system to rate the teaching performances.

Although the new assessment was a "remarkable step beyond multiple choice," Mr. Shulman said, "there still wasn't enough progress there."

In January 1988, Mr. Shulman and his associates turned for the next phase of their project to the development of highly structured portfolios to show the work of a teacher and class of students over time.

Twenty teachers agreed to construct portfolios during the 1988-89 school year. The documentation explored how to teach literacy skills to elementary students and introductory biology to high-school students.

Each teacher was given a set of instructions regarding the form of the portfolio and procedures to follow, and was asked to complete five of seven sections.

The candidates were given considerable latitude in determining the substance of their entries.

The biology teachers, for example, were asked to document how they would use the laboratory for an exercise. Examples of students' work were included in the portfolios.

Rather than ask each teacher to construct a portfolio alone, the Stanford researchers required that the portfolios be "coached" by school colleagues. Mr. Shulman compared the process to a doctoral program, where candidates' work is constantly evaluated and critiqued by professors.

The teachers turned in the portfolios by mid-May of this year for review and grading. Then they visited the assessment center in mid-June to discuss the work and their students' progress in formal interviews.

The research, Mr. Shulman said, showed that "it can be done. It is possible to break free from the limitations of testing."

During the coming year, he will be on sabbatical from Stanford. His colleagues, under the direction of project director Angelo Collins, will evaluate the data collected during both phases of the project and write a series of reports detailing their work.

Meanwhile, the administration of a consortium of states interested in developing new teacher assessments has shifted from Stanford to the offices of the Council of Chief State School Officers in Washington, D.C..

The Interstate New Teacher Assessment and Support Consortium is now being directed by Ramsay Selden, director of the council's State Education Assessment Center.

To date, 37 states are members of the consortium. Mr. Selden said invitations have been extended to all 50 states to join.

A new fee structure has been designed with the aim of making the consortium self-supporting. Each state will pay $2,000 to become a member, while institutions will be charged $1,000 and individuals will pay $300, Mr. Selden explained.

The consortium will present three conferences each year, publish a newsletter, and act as a clearinghouse to share information.

Web Only

You must be logged in to leave a comment. Login | Register
Ground Rules for Posting
We encourage lively debate, but please be respectful of others. Profanity and personal attacks are prohibited. By commenting, you are agreeing to abide by our user agreement.
All comments are public.

Back to Top Back to Top

Most Popular Stories

Viewed

Emailed

Recommended

Commented