Sharp increases in National Assessment of Education Progress (NAEP) math scores since 1990 may reflect little gain in student ability, a new report finds. The report, released last week by the Brown Center on Education Policy at the Brookings Institution, reveals that NAEP test questions do not incorporate much challenging arithmetic. In fact, the study reports, questions designed for eighth graders typically require only second- to fourth-grade arithmetic skills.
The report, How Well Are American Students Learning?, gauges the average arithmetic demand of NAEP from a sample of more than five hundred math questions from NAEP exams for fourth and eighth graders. It found that more than 40 percent of math questions addressed skills taught in first or second grade. (The questions are available to the public on the NAEP website, athttp://www.nces.ed.gov/nationsreportcard/ITMRLS/search.asp?picksubj=Mathematics.)
According to Tom Loveless, director of the Brown Center on Education Policy, too many of the questions rely on whole numbers, with few problems that involve fractions, decimals, and percentages. “Such instruction sets students up for trouble in more advanced high school classes and in daily life, where tasks such as shopping and measuring rarely involve neat, round numbers,” the Boston Globe quoted him as saying.
The report groups the questions by the conceptual area they test. On both the fourth- and eighth-grade exams, questions test general problem-solving ability or one of five specific “content strands”-number sense, measurement, geometry, data analysis, and algebra. Based on the level of arithmetic skill that questions in each area require, the report determines the average “arithmetic demand” of each section. As a measure of demand, the Singapore math textbook program-widely regarded as a challenging curriculum-is used. The report maintains that using most states’ grade-level standards would yield similar conclusions. According to Sharif Shakrani, deputy executive director of the National Assessment Governing Board, which sets the test content, the Singapore model leads to skewed results.
While NAEP suggests that today’s eighth graders know as much math as tenth graders did in 1990, score gains “only apply to the mathematics found on the NAEP test.” The report does not examine differences that may exist between NAEP exams for different years. Loveless said that it is unclear whether increased scores reflect true gains in math knowledge-especially since higher scores have not translated into larger enrollments in higher level math classes.
The report also finds that most middle school math teachers have little training in mathematics or mathematics teaching. Based on a random sample of 252 middle school math teachers nationwide, the report finds that less than 25 percent of America’s sixteen thousand middle school math teachers have a bachelor’s degree in mathematics. Another 25 percent of the sample had completed fewer than four college mathematics courses. More mathematics teachers-54 percent of eighth-grade teachers-earned degrees in education. Yet the teacher education level of sample math teachers was not impressive. Only 41 percent of the sample held a teaching certificate in mathematics.
The complete report is available at http://www.brookings.edu/GS/brown/bc_report/2004/2004report.htm.
“Study Finds National Math Test Easier” is available athttp://www.boston.com/news/education/k_12/articles/2004/11/18/study_national_math_test_easier/