Accountability and Testing
Recommendation
The state's end-of-year and end-of-course tests should be replaced with an independent, field-tested, and credible national test of student performance, such as the Iowa Test of Basic Skills, the California Achievement Test, or the Stanford 10 exam.
Background
State Board of Education officials recently admitted that, from the beginning, they set low standards for the state accountability program, the ABCs of Public Education, in order to gain support for comprehensive statewide testing. Unfortunately, since the program's inception in 1997, state officials never satisfactorily raised standards.
The Need to Change How Students are Assessed
• There has been a history of problems with our testing program. In 2001, state education officials were embarrassed to discover their new math tests were absurdly easy to pass — a product of poor judgment and a flawed system of field-testing. In addition, statistical problems with the 6th-grade reading scores in 2004 and 2005 prompted the State Board of Education to approve changes to the growth formulas used by the Department of Public Instruction to predict the performance of students from one grade to the next.
• Math standards are not demanding enough. While the 2005-06 revised math standards are an improvement over past years' standards, they are hardly rigorous. To be classified as proficient on the 2004-05 end-of-grade math tests, students were required to answer an average of 40.7 percent of the questions correctly. To be classified as proficient on the 2005-06 end-of-grade math tests, students were required to answer an average of 49.4 percent of the questions correctly.
• Guessing can vastly improve the chances that a student passes a state test. Students could be classified as proficient on the 2005-06 math tests if they knew the answers to just 32.5 percent of the questions (17 out of 50) and guessed on the rest. That percentage dropped to an average of 24 percent (12 out of 50) for students who could eliminate one of the four multiple choice test options before guessing.
• North Carolina's standards do not square with federal standards. While a high percentage of North Carolina's students are categorized as proficient on state tests (between 85 and 93 percent in 2005), a low percentage of North Carolina's students are categorized as proficient on the federal National Assessment of Educational Progress tests (between 32 and 40 percent in 2005).
• The state testing program lacks transparency. The Department of Public Instruction has conducted their testing program in near absolute secrecy. Past editions of the tests are not made public.
• North Carolina's testing results are confusing. Each school's results include the following terms and measurements: "growth designation," "ABCs status," "average growth," "achievement level," "c-ratio," and "performance composite." Parents must read a nine-page background packet just to understand what those terms and designations mean.
• North Carolina is one of only six states that oversee all aspects of state testing. Forty other states have contracts with testing companies, and 18 other states use an off-the-shelf/norm-referenced test.
Analyst: Terry Stoops
Education Policy Analyst
919/828.3876 • tstoops-at-johnlocke.org
|
