Share this

Imposing Tests on Teachers Would be Costly Error

May 12, 2010
Thomas Switzer- Monday,
Des Moines Register
December 4, 2000

Thomas Switzer, Dean, College of Education, University of Northern Iowa

There is great debate today about how best to guarantee that we have quality teachers for Iowa’s schools. Two pay plans suggest that substantially increasing the salary paid to Iowa teachers would both increase the numbers and the quality of those who enter the field. While increasing pay is certainly an important component in attracting and retaining a quality teaching force, it is not, however, the entire answer.

Some states have implemented extensive testing of teachers as a condition for entry into the profession in an attempt to improve the quality of their teaching force. One of the pay plans for Iowa also calls for testing of teachers as they progress through the pay grades. Such testing is, however, extremely costly and ineffective as an indicator of teacher quality. There are several reasons why this is true.

First, teachers are typically tested on low-level content with limited emphasis on teaching skills. Of course, we want all teachers to be competent in the content they teach. The tests, however, do not give us that confidence. The way teachers must know their content to be effective, especially at the high school level, is much more sophisticated. Effective teachers know how to extract from the knowledge base of their discipline those guiding principles that have explanatory power and are, therefore, most worth teaching. Knowing that a teacher understands content at this level of sophistication requires a much more elaborate form of assessment.

Second, since the typical form of testing does not give us the type of information we need to determine a teacher’s in-depth knowledge, it is not surprising then that there is no relationship between success on teacher tests and classroom performance of teachers. A high score on the test does not mean the person will be a good teacher. At a presentation by perhaps the major testing company in the country, I asked the question, "Is there a relationship between success on the teacher test and classroom performance?" I was told that no testing company would make that claim.

Still, with no proven relationship between test success and classroom performance, the data are often reported as though there were some real value. States report their test data with either great glee or great disdain, depending on the results. The tests seem to give the public—and especially politicians—a false sense of confidence that they have done something to improve teacher quality when, in fact, they have done nothing.

A 1998 bill passed by the U.S. Congress requires states that impose testing as a condition of teacher licensure to report that data to the U.S. Secretary of Education. The states are also required to rank their teacher education programs from top to bottom based on those test scores, a misuse of test data that in effect has no meaning. Iowa, as a non-testing state, is working to avoid this requirement.

What is needed, of course, is a much more sophisticated form of assessment that provides a level of confidence that a teacher can perform well in the classroom. At the University of Northern Iowa we are working on two initiatives to provide us with that confidence.

First is a project calling for teachers in training to present direct samples (a portfolio of sorts) of their actual performance in their classroom to a team of experts for evaluation. The National Board for Professional Teaching Standards [NBPTS] uses a similar process to assess the classroom performance of experienced teachers seeking national certification. Support for these teachers working toward NBPTS certification is provided by UNI. These two initiatives and their rigorous assessment are much more likely to provide the assurance that we will have quality teachers in the schools of Iowa.

It is likely that there will be pressures for the State of Iowa to impose a test for teachers as a condition for initial licensure. In my opinion that would be a costly mistake. Iowa has a reputation for high-quality schools and for preparing excellent teachers. We should not fall prey to the low-level testing being used in some states. I would hate to see the tax money of Iowa wasted on such meaningless testing. We know how to do it better in Iowa and should stick to our high standards. We should use assessment programs for teachers that give us the confidence that Iowa teachers can perform well in the classroom and that they can enhance student learning.