New standardized tests will be implented for fifth, sixth, and seventh grades
By: William Wichert
The state’s standardized testing system is changing, but the feeling among local school officials is not all fanfare and fireworks.
Some have said the proposed changes will give districts the useful information they need to improve instruction, but others are still concerned about remaining flaws in a formula that determines students’ passing rates every year.
The state Department of Education (DOE) announced last week that it was revising this formula to include a new standardized test for fifth, sixth, and seventh grades. Standardized testing currently takes place on the third-, fourth-, eighth- and 11th-grade levels only.
As required under the federal No Child Left Behind Act (NCLB), an interim version of the new test will be administered next spring, while a new advisory panel develops more broad changes to the overall testing and assessment system.
This new system will provide for a more detailed analysis of students’ results on each standardized test, and allow district officials to trace the progress of individual students across different grade levels.
"The state spends millions of dollars on these tests," said Acting Commissioner of Education Lucille E. Davy in a Nov. 16 prepared statement. "We must receive more education value from this investment and not simply use the tests to comply with federal reporting requirements."
Local school officials said they are still looking for the value of such tests.
Officials said they do not ignore the tests, but the results that come back from the state do not serve much of a practical purpose. Rather than show how students performed on specific questions, the results only show overall performance in general academic areas.
"That’s been the one complaint that has the most validity," said Northern Burlington County Regional Superintendent James Sarruda.
Dr. Sarruda stressed that while the district does not structure its curriculum in order to achieve high scores on standardized tests, the meaning of those scores is undermined because they only reflect how one individual class performed.
By the time the district receives test results from the state, the students who took those tests are already moving on to the next grade level, making it difficult for the district to apply those results to the new group of students.
"What’s the purpose of us all (school districts) taking a standardized test, so we can compare each other?" said Dr. Sarruda. "To create individual programs and portfolios, and track students through the grades, makes sense."
While state officials said they have yet to select a standardized test to administer to fifth-, sixth- and seventh-graders, North Hanover Superintendent Richard Carson said they should consider the TerraNova standardized test, which is taken annually by all K-6 students in his district.
This test, which is common among school districts near military bases, picks up where the state tests fall short by providing detailed analysis of how each student performed on each individual question.
"It’s just more valuable," said Dr. Carson. "Having these results, we use it to identify kids’ basic skills."
While the state may change how test results are reported, Bordentown Regional Superintendent John Polomano said the state should be looking at a deeper flaw within its testing system: the subgroup.
The subgroup is a group of students, based on race, economic status, and other factors, whose test scores can determine whether or not schools meet the NCLB-required Adequate Yearly Progress.
The formation of these subgroups and their role in testing results is a federal requirement, but each state decides how many students make up a subgroup. In New Jersey’s case, a subgroup is at least 20 studentsan arbitrary number that Mr. Polomano said improperly affects each school’s overall NCLB rank.
"(State officials) have no scientific basis for developing a subgroup," said Mr. Polomano. "The whole concept of how they’re implementing testing is flawed."
According to recent report by the New Jersey Association of School Administrators (NJASA), New Jersey is only one of about 10 states with subgroups of less than 30 students, making those test scores less reliable in determining the quality of an entire school.
"It is self-evident that a subgroup that is too small runs a substantial risk of identifying groups as failing to make AYP on the basis of chance rather than actual lack of performance," according to the report. The NJASA recommends in its report that the state increase its subgroup size to 50.
In a "frequently asked questions" section of its Web site, the DOE states that the subgroup size of 20 was determined to "best meet the needs of children while maintaining the highest level of certainty regarding a school’s proficiency level based on assessment tests.
"This combined approach allows for an increase in the probability of correctly identifying a school as being in need of improvement by holding the error rate (of 5 percent) constant to ensure the same level of accuracy for all groups being assessed," according to the DOE.
While some school officials may point to the flaws in the state’s testing system, Dr. Sarruda said the DOE’s proposed changes will at least give districts more detailed information about student performance on standardized tests.
"I am looking forward to the changes, because the system, right now, is not a very strong system," he said.

