Thursday, July 7, 2011

Holland Urges Careful Use of Test Scores to Evaluate Schools

School performance measures should not rely solely on tests that do not measure individual student improvement, state Rep. Corey Holland said today.
Rep. Corey Holland
“Test scores are very important and Oklahoma students need to demonstrate a certain level of knowledge,” said Holland, a Marlow Republican who is a former teacher. “But state-mandated test scores are just one piece of the education puzzle. Let’s not lose sight of that reality.”

Holland’s comments follow a massive cheating scandal in Atlanta, where an investigation found that 178 educators provided test answers to students or changed test answers to improve scores.

“The recent discovery of widespread cheating by teachers and administrators in Georgia is a wake up call to all of us,” Holland said. “These adults have been altering tests for years, and the cheating was not limited to just a handful of individuals. Over 100 people were involved, and many of them had previously been heralded as education heroes for the great improvements in their test scores over the years. If Oklahoma focuses solely on test scores we are not getting the complete picture of student achievement either. In Atlanta many assumed the students were doing much better than they actually were. Conversely, I fear many in Oklahoma assume our students are doing much worse than they actually are.”

Holland said Oklahoma tests currently do not provide adequate measurement of student learning and warned against making those tests the sole measurement of schools’ success.

“Oklahoma’s current testing system can compare one year’s class of students to last year’s class, but it currently fails to adequately measure whether a particular student has achieved a full year’s education growth during a school year,” Holland said. “Instead, the test measures if students are at grade-level without making reference to each student’s education status at the start of the year. This is inadequate data. High-stakes testing can make the situation worse by making the focus entirely on a defined level the average student in a given grade should reach.

“Not all development in students occurs at the same rate. Some students do well in some areas early, while others develop a little later. We have all seen this in little league. Does every little league hero wind up being the best in high school? Not often. Academic development often involves the same differences in timing between students as they progress.”

Unless the testing system is improved to provide true measurement of student performance, Holland said relying on test results as the major determining factor in funding and policy implementation could have significant unintended consequences like those seen in Atlanta.

“What occurred in Georgia is inexcusable. The administrators and teachers in Atlanta orchestrated a scheme to defraud their students, parents and the public, and they should be held responsible for their actions,” Holland said. “The lesson to learn is that overemphasizing one aspect of the education process is a mistake. Oklahoma needs to develop a statewide longitudinal data system that takes into account all the learning occurring in a classroom. As education leaders we have a responsibility to make our decisions based in reality. We should also want to measure a student’s total achievement during a school year. A test score is a snapshot in time. The people in Atlanta can now testify how test scores alone don’t always reflect reality.”
Related Posts Plugin for WordPress, Blogger...