This op-ed appeared in the Newark Star-Ledger on Sunday, May 19.
By Barbara Keshishian
New Jersey teachers and parents are anxious about the state’s new evaluation procedures, and the Department of Education isn’t helping.
Last year, the New Jersey Education Association helped craft a new tenure law that reduced the time and cost of dismissing teachers deemed to be ineffective, while ensuring fairness.
The law also mandates a new teacher evaluation process, intended to foster excellent instruction and expanded student learning.
But using student standardized test scores to gauge teacher effectiveness, part of a controversial national “corporate” reform movement strongly supported by the Christie administration and its Department of Education, has raised real concerns among educators and evaluation experts.
Two years ago, 13 stakeholders — including NJEA and Educational Testing Service, which designs tests — sponsored a symposium on the subject at ETS’s Princeton campus.
The unanimous conclusion of a panel of experts was “not so fast.” Two-thirds of the factors affecting test scores occur outside the classroom (in the home and community), and ETS concurred with that concern and several others in a policy publication distributed at the symposium.
Outside factors like family circumstances, poverty, parental involvement, and English language proficiency affect test scores — and are beyond teachers’ control.
Test score data can be wildly misleading. New York City tracked 18,000 teachers from 2007 to 2010, rating them from 1 to 100 in improving student test scores. Some teachers had an 80-point swing in improving test scores from one year to the next.
Hundreds of elementary teachers were rated both “highly effective” and “highly ineffective” in the same year, because their students scored well on one subject test but poorly on another. Whatever those tests measure, it’s clearly not teacher effectiveness.
There are two very different models for measuring test score data — “Value-Added” models and “Student Growth Percentiles.” VAMs measure changes in a student’s performance from year to year, while SGPs consider the performance of each student in relation to similar students. SGPs are used by the Department of Education.
Research on VAMs is inconclusive, but at least they try to take into account the aforementioned outside factors.
But SGPs are less reliable, because they don’t. Even their designer, Damien W. Benebetter, notes that SGPs are not intended to make high-stakes personnel decisions.
Now, the Department of Education is sending mixed messages.
Last November, at a roundtable in Newark, the Department of Education’s director of evaluation, Timothy Matheny, was asked to address the fact that the science on using test data to evaluate teachers “isn’t great.”
“We do hope the science gets better,” replied Matheny, as the audience laughed. He added: “We’re taking into account all those arguments and concerns raised about (VAMs), for example, and that’s one reason we’re looking at (SGPs) as the appropriate approach.”
But Matheny’s boss, Commissioner of Education Chris Cerf, keeps citing another study based on VAMs — ignoring the SGPs that New Jersey uses. Who’s on first?
That disagreement didn’t prevent this outrageous comment in a May 1 Star-Ledger story: “Cerf said research shows test scores are ‘far and away’ the best gauge of teacher effectiveness, and to not use test score data would be ‘very anti-child.’ ”
In fact, there is no research showing that test scores are “the best gauge of teacher effectiveness,” and Cerf’s use of a VAM-based study to push SGPs is a classic example of comparing apples to oranges.
While other states are considering jumping off the testing bandwagon, New Jersey wants to test every subject in every grade, the research be damned. Teachers want to work in collaborative settings that foster their best work, but that’s not what the Department of Education is creating.
Now would be a good time to take a deep breath, study the research, and create an evaluation system to ensure the best teachers will be in every classroom — a goal the NJEA fully supports and endorses.
Barbara Keshishian is president of the 200,000-member New Jersey Education Association.