Monday, March 7, 2011

SB736/HB7019: The trouble with value-added measurement

The trouble with value-added measurement is

It's not on the test

SB736/HB7019 intends to apply value-added measurement (VAM) to connect student achievement on a test as a performance indicator of the teacher, or a poor student score = poor teaching. VAM uses statistical tools to calculate the contributions a teacher makes to student achievement gains. The formula for the calculation includes factors the developer chooses to include. What do experts have to say about the value of value-added measurement?

1) Jim Angermeyr, Director of Research, Evaluation & Testing for Bloomington Public Schools and "one of the designers of a widely respected value-added test lots of Minnesota schoolchildren take two or three times a year" was interviewed by Beth Hawkins. In the article, "Do 'value-added' teacher data really add value?", Angermeyr is described as "something of a standardized testing skeptic. He believes that economists tend to believe in using value-added data in evaluation. Educators and psychometricians, not so much."
“It’s not necessarily that the methodologies are wrong,” he said. “It’s that the inferences we’re drawing can be wrong.”
"The kids are the greatest of the variables, of course. The tests may tell you a student is reading better or sliding in math, but they don’t tell you whether she spent the summer with a tutor or he is so young the test isn’t as accurate as it would be in an older child.

Nor is the same test used from year to year. A particular student or teacher may fare better on a test closely normed with curriculum vs. one aligned with a set of knowledge-based standards."

You leave out a lot of the potential variables,” Angermeyr said.  “They’re just not at the point where we should use them to make decisions about jobs.”
2) The National Research Council and the National Academy of Education gathered experts in the this field and held a workshop titled "Getting Value Out of Value Added." There was general agreement by participants that there is no single statistical model and still a work in progress for applicability when measuring teacher quality.

Discussion resulted in these additional conclusions:

Results generated by existing models have a high degree of instability.

Good results require good tests and good test results.

Experts find it useful for low stakes use to identify areas that need improvement.

Experts recommend against using it for high stakes such as, teacher pay.

According to Henry Braun, the group determined the following:"
To nobody’s surprise, there is not one dominant VAM. Each major class of models has shortcomings, there is no consensus on the best approaches, and little work has been done on synthesizing the best aspects of each approach. There are questions about the accuracy and stability of value-added estimates of schools, teachers, or program effects. More needs to be learned about how these properties differ, using different value-added techniques and under different conditions. Most of the workshop participants argued that steps need to be taken to improve accuracy if the estimates are to be used as a primary indicator for high-stakes decisions; rather, value-added estimates should best be used in combination with other indicators. But most thought that the degree of precision and stability does seem sufficient to justify low-stakes uses of value-added results for research, evaluation, or improvement when there are no serious consequences for individual teachers, administrators, or students." (p.54)

While Florida plunges into creating dozens of new tests, North Carolina's legislature in bipartisan agreement is sending a bill to the Governor to end some end-of-course tests, maintaining those to meet federal requirements and to measure student achievement. Two years ago, they voted to end a few others. This legislature wishes to stop paying for so many tests that are both expensive and failing to yield the returns once thought beneficial.

Concerns over the costs for implementing SB736/HB7019 by Florida Senators and by school boards are being reported. In the end, this is an unfunded mandate using an experimental statistical model, and an expensive test development and implementation scheme that extends far beyond the reach of RT3 dollars. Legislators remain silent on the issue of costs and cost benefits. The public has a right to know.

To read the report, access the widget on Grumpy Educators located on the right size of the page. Select the open book with the word Read underneath, then select Open Book on the small screen. The document  will open in an easy to read size

Top Read all of Sandra in Brevards education blogs visit Grumpy Educators


The experts in the National Research Council and the National Academy of Education in this conference included:

Rita Ahrens, Education Policy Studies

Joan Auchter, National Board for Professional Teaching Standards

Terri Baker, Center for Education, The National Academies

Dale Ballou, Vanderbilt University

Henry Braun, Boston College

Derek Briggs, University of Colorado at Boulder

Tom Broitman, National Board for Professional Teaching Standards

Alice Cain, House Committee on Education and Labor

Duncan Chaplin, Mathematica Policy Research

Naomi Chudowsky, Center for Education, The National Academies

Pat DeVito, AE Concepts

Beverly Donohue, New Visions for Public Schools

Karen Douglas, International Reading Association

Kelly Duncan, Center for Education, The National Academies

John Q. Easton, Consortium on Chicago School Research

Stuart W. Elliott, Center for Education, The National Academies

Maria Ferrão, Universidade da Beira Interior, Portugal

Rebecca Fitch, Office of Civil Rights

Shannon Fox, National Board for Professional Teaching Standards

Jianbin Fu, Educational Testing Service

Adam Gamoran, University of Wisconsin–Madison

Karen Golembeski, National Association for Learning Disabilities

Robert Gordon, Center for American Progress

Jeffrey Grigg, University of Wisconsin

Victoria Hammer, Department of Education

Jane Hannaway, Education Policy Center

Patricia Harvey, Center for Education, The National Academies

Lloyd Horwich, House Committee on Education and Labor

Lindsey Hunsicker, Committee on Health, Education, Labor, and Pensions

Ben Jensen, Organisation for Economic Co-operation and Development

Ashish Jha, Harvard School of Public Health

Moshe Justman, Ben Gurion University, Israel

Laura Kaloi, National Association for Learning Disabilities

Michael Kane, National Conference of Bar Examiners

Judith Koenig, Center for Education, The National Academies

Michael J. Kolen, University of Iowa

Adam Korobow, LMI Research Institute

Helen F. Ladd, Duke University

Kevin Lang, Boston University

Sharon Lewis, House Committee on Education and Labor

Valerie Link, Educational Testing Service

Dane Linn, National Governors Association

Robert L. Linn, University of Colorado at Boulder

J.R. Lockwood, RAND Corporation

Angela Mannici, American Federation of Teachers

Scott Marion, National Center for the Improvement of Educational Assessment

Daniel F. McCaffrey, RAND Corporation

Alexis Miller, LMI Research Institute

Raegen Miller, Center for American Progress

John Papay, Harvard University

Liz Potamites, Mathematica Policy Research

Ali Protik, Mathematica Policy Research

Sean Reardon, Stanford University

Mark D. Reckase, Michigan State University

Andre Rupp, University of Maryland

Sheila Schultz, HumRRO

Lorrie Shepard, University of Colorado at Boulder

Judith Singer, Harvard University

Andrea Solarz, Director of Research Initiatives, National Academy of Education

Gerald Sroufe, American Educational Research Association

Brian Stecher, RAND Corporation

Justin Stone, American Federation of Teachers

David Wakelyn, National Governors Association

Greg White, Executive Director, National Academy of Education

J. Douglas Willms, University of New Brunswick

Mark Wilson, University of California, Berkeley

Laurie Wise, HumRRO


  1. It would come as a big shock to some very ifluencial people if Americans started questioning the emphasis we have been putting on standardized tests.

  2. Shouldn't we test the teachers?

  3. Shouldn't we test people before they have children or just grade them after kids are born?
    I think we should start testing the water in Tallahassee first and then give tests to all the staffers in Florida's Ed department, and all legislators to be sure they can add and think.
    Quietly we should buy stocks in the testing firms. Looks like they 'll be nice and fat for awhile. They have jobs for consultants and test scorers too.

  4. Madpole.. we need a better way to evaluate school and teacher perfomance.. AS it is now, we complain about teachers, teaching to the test. If half their yearly income depends on the outcome of a 3o or 60 minute test, preping kids for the test is all they'll do.

    Anonymous... been wondering about the water in Tally for a while. Some places it's safer to drink the wine than the water. Half loaded they might not muck things up as much. From what I've heard testing companies don't pay scorers too well.. If you've got some connections there's going to good money selling the tests and servicing the customers.


Note: Only a member of this blog may post a comment.