by Stuart Singer, The Teacher Leader

In a recent post I decried the decision by the Los Angeles United School District to publicly release a list ranking 6,000 elementary school teachers based on their students’ standardized test scores.  My main argument against this practice was the havoc it would cause in terms of teacher cooperation, staff morale and administrative anxiety.   My assumption at the time I posted this blog was that at least the standardized tests that the teacher evaluations were based on were valid – maybe not perfect, but valid.  However, recent disclosures about the validity of the New York Regents exams has caused me to question the suitability of using end-of-course test scores in any effort to evaluate teachers. 

The Gold Standard No More

During the implementation of the Virginia Standards of Learning (SOL) exams, teachers were instructed to use the Regents exams of New York as a model for what the state was trying to accomplish with their standardized tests.  Sample Regents or Regents-like exam questions were disseminated to assist teachers in preparing their students for the SOLs.  Virginia clearly wanted to develop a set of exams to demonstrate mastery of a subject in a manner similar to the New York exams. 

Flash forward to 2010.  A recent article in the Albany Times Union reveals that passing the Regents exams has little relevance to a student’s educational accomplishments.  As reported by Times Union education reporter Scott Waldman:

“A quarter of New York’s college students in two- and four-year schools need extra academic help, according to the Education Department. And though nearly three-quarters of students have passed the core Regents exams for the last three school years, just a third of them scored over 85, the bar set by SUNY schools.”

To complete the downgrading of a once proud educational innovation, Waldman adds that the Regents are now “so hard to fail they have become meaningless.” 

The teachers preparing students to take test understand that the rigor has been so reduced that it is now virtually irrelevant.  In order to graduate a student must score at least 65 on five Regents exams.  According to a social studies teacher in Queens a student can miss 15 of 50 multiple choice questions on one of the exams and still earn a raw score of 90. 

Tip of the Iceberg

How many of the tests being used by states to monitor student progress are being created poorly and graded ineffectually?  I have already shown through a statistical analysis that due to the construction of the test (multiple-choice with no penalty for guessing) and a low passing score (50%), an individual can pass the Virginia Algebra 1 SOL exam by answering slightly less than 40% of the questions correctly. These standards should not equate passing with demonstrating mastery of a subject.  Moreover, making the Regents, SOLs and any other barrier exams so easy that they become extremely difficult to fail should cast serious doubt about their reliability as a component of teacher evaluations.  If the Obama Administration, the LAUSD and others want to tie teacher performance to student test scores there needs to be a significant commitment to creating tests that accurately determine a student’s actual comprehension of the curriculum and a teacher’s ability to implement a program that delivers that knowledge.  That commitment will include the funding to write and grade tests that are not exclusively multiple-choice and the courage to establish standards that will reveal more accurately the success or failure of a school system.