Testing Teachers to Raise Standards: Does it Work?

by
Harold Berlak

Center for Education Research, Analysis, and Innovation
School of Education
University of Wisconsin-Milwaukee
PO Box 413
Milwaukee WI 53201
414-229-2716

May 23, 2000

CERAI-00-12

Testing Teachers to Raise Standards: Doesit Work?

By Harold Berlak

Parents, politicians andschool boards across the nation are rightly concerned about the quality ofteaching and learning in the nation’s elementary and secondary schools.

The nation’s largest citiesreport growing shortages of teachers who are both legally credentialed andfully qualified. The problem is especially acute in public schools that servethe poor, African-American, Latino, and immigrant communities. Many childrenare being taught by a continuous and changing procession of "permanentsubstitutes."

High turnover, especiallyamong entry-level teachers, plagues the profession. According to a recentreport of the National Commission on Teaching and America’s Future, anextraordinary 30 percent of new teachers -- after spending five or six years inpursuit of an educational credential -- leave in less than five years. ^[1] The exodus continues and is particularlystriking in mathematics, the sciences, computer technology, and otherspecialized curricular areas.

Teachers as a group areheld in low esteem, and in spite of the fact that they are highly unionized,their salaries fall further behind other public service employees andprofessionals with comparable years of education. ^[2]

As we enter the high seasonof election year politics, the pollsters tell us that education reform is atthe top of the list of public’s concerns. The remedies aspiring office holders,from the presumptive presidential candidates on down, offer in the face of suchconcerns are unfortunately a replay of the familiar.

At the center of almost everyproposal circulating today to raise the quality of teaching is more teachertesting. In recent months even the American Federation of Teachers has joinedin the call for a new national teacher test..

At first glance, a policyintended to weed out the illiterate and raise the quality of the pool ofcandidates for education careers appears to be a reasonable, common-sensesolution to the problem of low teacher quality. Who could possibly object?

Yet, in spite of honorableintentions, the tests that the vast majority of states now require as acondition for entry into teaching and other public school professions aremaking matters worse, especially for minority teachers. These mandatory teststhat limit who can enter the profession have failed on two counts: They havenot raised the standards of teaching and learning in classrooms, and they havesharply reduced the numbers of qualified teacher candidates -- especiallyso-called "minority" candidates.

The current wave of teachertesting began in 1983, when the California legislature, responding to the callfor educational excellence, imposed a basic reading, writing and math testcalled CBEST as a condition for entry to a teacher credential program. Todaythere are, according to the National Research Council, 41 states that requireprospective teachers to pass one or more tests. [^3]

These are not licensuretests equivalent to bar exams or medical boards, which are controlled by theprofessions and taken after completion of training. Instead they arehigh-stakes gatekeeper tests, used to restrict entry  toprofessional training. Unlike the SAT, GRE, or other admissions tests requiredby law or medical schools, however, the teacher tests are enshrined in legalmandates. Results on this single test override all other state requirements --including successful teaching experience.

Thirty-seven states mandatesuch a "basic skills" test. Others offer a mix of tests intended toassess pedagogical or subject matter knowledge, or both. Virtually all thesetests depend entirely on the familiar fill-in-the-bubble multiple choice testtechnology, sometimes augmented by a few standardized, open-ended questions.

This 17-year-old Californiapolicy has not achieved its avowed purpose of raising the literacy levels ofCalifornia’s educational professionals, however. Nor has CBEST contributed toenhancing learning and achievement among California’s children.

The evidence is found in alarge body of research and many thousands of pages of testimony and studiescommissioned by both sides of a civil rights suit brought against the state ofCalifornia and challenging CBEST. ^[4]

Not only has CBEST failedat its avowed purposes, it has greatly compounded an already serious shortageof teachers and school administrators and other specialists by unnecessarilyreducing the pool of competent, caring, and qualified educators. Moreover, by atwo-to-one margin, those denied were disproportionately persons of color --African- Americans, Asian-Americans and Latinos.

Some have argued that whilethis is unfortunate, it is the price we must pay for raising standards. Yetwhat is striking in this body of work is that there is no empirical evidence whatsoeverto support this claim.

Regardless of theorganization conducting the research, including the studies commissioned by theState of California itself:

· There is no connection between aperson’s performance on the tests and his or her performance as a teacher;

· There are no indications that thatteachers who successfully complete other requirements for a credential lackbasic literacy skills;

· The single greatest obstruction toincreasing the proportion of persons of color in the teaching force is CBEST. ^{[5], [6]}

A recently released interimreport of the National Research Council (Tests and Teaching Quality,2000) confirms that there are there is no significant connection between standardizedteacher tests and actual performance; that the tests have no demonstrablepredictive value; and that those who are disproportionately excluded by thetests from the teaching pool are persons of color.

Understanding why this isso requires a lengthy explanation beyond the scope of this brief commentary.Very simply, however, the test’s authors used widely discredited methods tovalidate it; California’s newly elected Superintendent of Public Instructionarbitrarily raised the test’s minimum passing scores far above the levelsproposed by the test’s authors; and the testing technology itself is based on anumber of implausible presumptions, one of which is that a fixed percentage oftest-takers must fail, regardless of their raw scores. Finally, the testignores the persistence of cultural and racial bias that continues to pervadethe technology of testing and the testing process. ^[7]

The use of standardizedteacher tests to control access to teaching as a career is self-defeating. Itcompounds the problems of achieving educational excellence with equity. Focuson testing has served to divert us from addressing the problems of providingevery child with the opportunity to be taught by a qualified, knowledgeable andcaring teacher. Because there is no connection of tests to job performance,furthermore, one of the chief consequences of the use of teacher testing,whether intended or not, is to is to install and to strengthen a form ofinstitutionalized racism.

Testing teachers as a fixfor raising educational standards is just one part of the education reformagenda advanced by Presidents Bush and Clinton and now embraced by those whoseek to be their successors. High-stakes testing across the school population,from the youngest students to the most senior teacher, has become the backboneof the nation’s prescription for improving our schools.

These policies, hatched bypoliticians and the established Washington think tanks, maintain the support ofcorporate leaders, governors, and many national and state legislators. Theywill not end of their own accord, but only when the public sees through thesham of raising educational standards by mandating tests, and demands someserious answers.

ENDNOTES

1. Darling-Hammond, Linda, Solving The Dilemmas of Teacher Supply,Demand, and Standards. National Commission on Teaching & America’sFuture. New York, August 1999. Available athttp://www.tc.columbia.edu/~teachcomm/CONFERENCE-99/SOLVING/

2. Education Week Vol. XIX no.18, Jan. 13, 2000, p. 36; citingU.S. Census data 1992-1999. Available at http://www.edweek.org/sreports/qc00/

3. Committee on Assessment and Teacher Quality, Tests and TeachingQuality, Interim Report. National Research Council, National Academy Press,Washington, D.C. Available at http://books.nap.edu/catalog/9788.html

4. Assn. of Mexican-American Educators et al v. State of California,Civil Rights Class Action C92-3874-WHO, U.S. District Court of NorthernCalifornia.

5. Poggio, John, Direct Testimony of Plaintiff’s Expert, Civil RightsClass Action C92-3874-WHO, U.S. District Court of Northern California, 1995

6. Izu, Jo Ann, C. Long, K. Stansbury, and D. Tierney, AssessmentComponent of the California New Teachers Project: Evaluation and AssessmentPractices, Far West Laboratory for Educational Research and Development.San Francisco, 1992

7. Berlak, Harold, Adverse Impact: How CBEST Fails the People ofCalifornia, Applied Research Center, Oakland, 1999. Available athttp://www.arc.org/Pages/Ecbest.html

cerai-00-12