Tests, Testing, and Genuine School Reform

Herbert J. Walberg

• Publisher: Hoover Institution Press
• Year: 2013
• Print ISBN: 9780817913540
• eBook ISBN: 9780817913564
• Edition: 1
• Topic: Education
• Subtopic: Education General

The author draws on scientific studies of tests and their uses to show how standardized achievement tests must play a central role in improving achievement in K-12 schools. He explains the central considerations in developing and evaluating tests and tells how tests can best be best used, covering such topics as using tests for student incentives, paying teachers for performance, and using tests in efforts to attain new state and national standards.

1

INTRODUCTION AND OVERVIEW

 

The pressing need to improve achievement in American schools is widely recognized. Factors associated with high achievement are appropriate testing, along with high standards, a curriculum closely adapted to the standards, and effective teaching. Because test design and use are technical matters, legislators, state and local school board members, and educators themselves are often poorly informed about their strengths, potentials, and limitations.

Despite good intentions, responsible officials often adopt misguided testing policies, and teachers have used tests that do not accomplish their intended purposes. For these reasons, the apparent results badly inform parents, citizens, and policy-makers about the actual achievement of students—a reason for American students’ mediocre performance relative to those in other economically advanced countries and relative to the new demands of the information economy.

This book draws on scientific studies of tests and their use to inform users and consumers about well-established principles of testing, current problems involving their use, and evidence-based solutions.¹ In addition, because valid tests cannot be developed without high and specific standards, one of the chapters discusses standards and how they should determine the plans and development of tests.

These topics are particularly important today. In the past several decades, costs of public schools have steadily and substantially risen, yet student achievement has remained stagnant. In response, schools, school districts, states, and the federal government are adapting a reform framework of “standards-assessment-accountability,” with achievement tests playing a central role in assessing what students have learned.

To perform their role in the reform framework, tests must be technically adequate and well administered. They must be aligned with standards and reported accurately and fairly to the interested parties, including parents, educators, school boards, legislators, and citizens. These contributions are all the more important for high-stakes decisions such as requiring failing students to repeat grades, closing or chartering repeatedly failing schools, and paying teachers for the achievement progress of their students, topics that are also discussed in this book.

Origins of Achievement Testing

The history of testing can be traced to the beginnings of tribal societies, when they were used to determine whether young people were ready to assume adult responsibilities. The kinds of achievement tests that are the focus of this book have a far shorter history. They evolved from attempts to use scientific methods to understand human intelligence. Those investigations began with psychological interests in differences in ability, emotions, and behavior among humans. In England, for example, Sir Francis Galton surveyed the abilities of British families that led to debates about whether differences in intelligence and human functioning are attributable to heredity or environment.² Intelligence testing grew out of these early investigations notably in France and the U.S. around 1905, which led testing specialists to generate measures for evaluating human potential. During World War I, the United States’ military used uniform tests and scoring for assigning personnel to jobs. By 1933, thousands of tests were in use for measuring intelligence, aptitude, and personality.³

After 1950 the emphasis in school testing began shifting from personality and potential to academic accomplishment or achievement in mathematics, reading, science, and other subjects and skills. By establishing common metrics for comparing achievement of individuals from all social classes, educational backgrounds, and cultures, standardized tests⁴ could show objective evidence of student progress, readiness for college and graduate and professional schools, mastery of English and math, and employment skills.

College admission tests led the growth of standardized testing. The non-profit Educational Testing Service, for example, was founded in 1947 to meet the needs expressed by the American Council on Education, the Carnegie Foundation for the Advancement of Teaching, and the College Entrance Examination Board. These organizations were founded on the belief that admission policies would be more fair if students from a wide range of academic, geographical, and social backgrounds—not just descendents of alumni and Eastern Seaboard families who tended to be admitted to prestigious Ivy League schools—were given an opportunity to compete on objective, standardized tests.

Standardized tests continued to gain importance. Federal initiatives such as the Elementary and Secondary Education Act of 1965 encouraged schools to use standardized achievement tests earlier in a student’s educational career to determine success in the usual school subjects. After 1970 objective examinations were used increasingly for occupational licensing and by firms to measure potential candidates’ knowledge and skills. Increasingly, individuals who passed high-stakes examinations could earn diplomas, receive scholarships, and obtain licenses to practice in professional fields such as law and medicine. Failure restricted these opportunities, but the test results could offer individuals information on how they might correct their deficits.

Criticism of Standardized Tests

This book focuses on standardized achievement tests in which all those tested face the same tasks and conditions. High scores mean students have acquired the knowledge and skills they need to meet increasingly important standards and ready themselves for further learning in school. Most individuals take standardized tests several times during their lifetime. Necessary tasks like obtaining a driver’s license require the successful completion of standardized tests, making them difficult to avoid; and tests are used routinely even in voluntary activities such as first-aid instruction.

Objections to Standardized Achievement Testing

Several influential education writers adamantly oppose current models of standardized testing and the growing emphasis on high standards and standards-based testing.⁵ Alfie Kohn, for example, urges educators to “make the fight against standardized tests our top priority until we have chased this monster from our schools.” Similarly, Gerald Bracey holds that “high standards and high-stakes testing are infernal machines of social destruction.”

Though more tempered, several political leaders have also expressed misguided criticism about standardized achievement tests and asked policy-makers and educators to avoid them. Before he became president, for example, Sen. Barack Obama urged “innovative assessments, including digital portfolios,” and making “the goal of educational testing the same as medical testing—to diagnose a student’s needs,”⁶ leaving out educator and student accountability for learning. This book makes clear why critics of standardized testing are wrong and how their views, if acted on, would undermine learning. Similarly, state legislators have allowed lax standards and ill-conceived tests to measure the progress of schools, educators, and students.

Tests as Guides to Policy

For another reason, high standards, valid tests, and accountability are important for America’s future. The United States has traditionally excelled in adult accomplishments in mathematics and science as well as their practical applications, but this status is now threatened. Although top American universities are second to none in the world, the National Science Board reported that the U.S. lead is shrinking.

Foreign students, moreover, comprise an increasingly larger percentage of students in American university graduate programs in these scientific and technical fields. They often return home with the best training American universities offer. Many American students are unable to show similar levels of achievement.

Countries in Asia and Europe, moreover, have increasingly improved primary and secondary education, which may be even more decisive in scientific and technical leadership. Long before the school achievement crisis was recognized, I pointed out that the United States’ welfare and prosperity benefit more from a well-educated population than from a scientific elite making scientific discoveries. Why? Credible scientific discoveries are published in peer-reviewed journals easily accessible outside the country of origin.⁷ Undue efforts to discover scientific breakthroughs, moreover, can divert time and energy from making effective use of them in such fields as medicine and engineering. The United States, for example, leads the world in medical research but lags behind other countries in children’s health and adult life expectancy.

The importance of high standards and attaining better test scores in schools is better recognized now than when I was writing in the early 1980s.⁸ During the past few decades, tests have been employed to measure how well K–12 students and schools have met proficiency standards. Tests and standards have become more effectively employed to advance student knowledge and skills. Analysis of K–12 standards and tests, however, reveals continuing problems.

Achievement tests today play a major role in K–12 education. They allow educators to assess the progress of students, identify their strengths and weaknesses, and plan remediation as well as revisions of teaching and curriculum. They allow parents to objectively monitor their children’s learning and similarly intervene to help solve their learning problems. They enable education leaders to monitor and evaluate the progress of schools and individual teachers. Achievement tests also help citizens and legislators inform themselves about the efficiency of the large amounts of tax money spent on schools.

Surveys show that parents, citizens, and legislators strongly support tests and testing and want to see bigger consequences for excellence and failure. Educators—particularly professors in schools of education where teachers and administrators are trained—unfortunately lack testing expertise and often oppose their use. Until recently, educators prevailed. But the growing awareness that American public schools are failing to fulfill their responsibilities has changed the terms of the debate, and educators are increasingly being held accountable for disappointing achievement results.

The following chapters present much research supporting the positive and substantial effects of standardized tests. The first four chapters of this book explain the most important ideas about achievement tests and the steps in developing good tests. The reader is shown why tests are necessary, how to recognize well-made tests, and how tests are properly written. The next four chapters turn from tests to testing, that is, how tests can be best used. These include such topics as using tests to motivate students and teachers for better performance, how to prevent test fraud by students and teachers, and the role of tests in meeting state and national standards. In view of the continuing controversy over test design and instances of fraud, the final chapter argues that the development, administration, and scoring of tests and reporting of results should be conducted by organizations independent of traditional school authorities. A brief conclusion summarizes the key findings.

 

 

1. Many books concern testing, such as F. Allan Hanson’s Testing, Testing: Social Consequences of the Examined Life (Berkeley, CA: University of California Press, 1993) and Williamson M. Evers and Herbert J. Walberg, eds., Testing Student Learning, Evaluating Teaching Effectiveness (Stanford, CA: Hoover Institution, 2004). The Brookings Institution, Fordham Foundation, Heartland Institute, Hoover Institution, and other organizations offer policy commentary on standardized testing.

2. Sir Francis Galton, Hereditary Genius (London: MacMillan Publishing, 1869). In much of education, achievement tests largely and constructively replaced intelligence testing since they indicate actual accomplishment, a more solid basis of remediation and further instruction, and prediction of further achievement in the specific area tested.

3. Gertrude H. Hildreth, Bibliography of Mental Tests and Rating Scales (New York: The Psychological Corporation, 1933).

4. A test is standardized if it is given under uniform conditions so that a student’s resulting test score can be compared to those of other students in norm-referenced testing or a performance criterion in criterion- or standards-based testing. It is a regrettable precedent in testing that the similar terms “standard” and “standardized” are easily confused, but it can be hoped that the different meanings of these similar terms are clear in context.

5. Alfie Kohn, “Burnt at the High Stakes,” Journal of Teacher Education, 2000, 51, 349; Gerald Bracey, “International Comparisons: An Excuse to Avoid Meaningful Educational Reform,” Education Week, January 23, 2002, 24.

6. Barack Obama, “Our Kids, Our Future,” November 20, 2007, remarks, Manchester, New Hampshire. http://usliberals.about.com/od/education/a/ObamaEdPlan_4.htm

7. Herbert J. Walberg, “Scientific Literacy and Economic Productivity in International Perspective,” Daedalus, 1983, 112, 1–28.

8. See, for example, John Cronin, Michael Dahlin, Yun Xiang, and Donna McCahon, The Accountability Illusion (Washington, DC: Thomas B. Fordham Foundation, February 2009); John Cronin, Michael Dahlin, Deborah Adkins, and G. Gage Kingsbury, The Proficiency Illusion (Washington, DC: Thomas B. Fordham Foundation, October 2007); Chester E. Finn Jr., Liam Julian, and Michael J. Petrilli, The State of State Standards 2006 (Washington, DC: Thomas B. Fordham Foundation, August 2006).

2

WHY TESTS ARE NECESSARY

Well-constructed standardized tests can help us assess how well students achieve broad, commonly valued academic goals. By evaluating the test performance of groups of students at the same grade level or who have taken the same course, and comparing it with a representative range of students from that group, it is possible to determine how well schools are fostering academic achievement.¹

Standardized tests can measure the degree to which students attain proficiency standards for specific topics and grade levels set by state and national governments. Standardized tests also make it possible to compare students in one school system, city, state, or country to those in others systems and places, which can reveal insights on what kinds of educational practices work best and which workforce is best prepared to compete in the global economy.²

By designing tests with common content, directions, and scoring procedures and administering these tests under the same conditions, standardi...

Table of contents

• Cover
• Title Page
• Copyright Page
• Contents
• Acknowledgments
• 1. Introduction and Overview
• 2. Why Tests Are Necessary
• 3. Well-Made Tests
• 4. Types and Uses of Achievement Tests
• 5. Preparing Standardized Achievement Tests
• 6. Tests as Incentives
• 7. Preventing Test Fraud
• 8. Standards and Testing
• 9. Using Tests to Raise Student Achievement
• 10. Conclusion
• About the Author
• About the Hoover Institution's Koret Task Force on K-12 Education
• Index