Ohio can't rely on do-it-yourself academic accountability
As lawmakers in Washington hash out the details of the reauthorization of No Child Left Behind (NCLB), both accountability and standardized testing are facing mounting criticism and skepticism. This backlash is felt in the Buckeye State, where some would like to move our academic accountability system away from the state's current achievement tests. While many speak of national standards and a national assessment (Fordham included), there are powerful forces in Ohio encouraging us to take a step backward instead.
Governor Strickland favors multiple measures of performance as the basis for academic accountability (see here). Teachers and principals participating in the administration's education focus groups say that portfolio assessments are a major topic of discussion. And George Wood, author of Strickland's K-12 education transition report, is a critic of the standards-and-testing movement. On his blog, Wood writes about Nebraska's assessment system as a model to which Ohio and other states should look for guidance and inspiration (see here).
Nebraska's STARS (School-based, Teacher-led Assessment and Reporting System) allows the state's 517 school districts to design their own assessment systems. These assessments include a portfolio of teacher classroom assessments, district tests that measure how well children are meeting locally developed learning standards, a state writing test, and at least one nationally standardized test to serve as what Wood calls a "reality check" (see here).
Results of these local assessments, not surprisingly, show that Nebraska's students are making academic gains (see here). Wood wonders, then, in a recent entry why the Cornhusker State's legislature decided to mess with a good thing last summer by enacting the state's first statewide standardized assessments in reading and math. One guess is that legislators looked at the gold-standard national "reality check" and questioned how well the state's students were actually performing.
On the 2007 National Assessment of Educational Progress (NAEP), only 38 percent of Nebraska's fourth graders scored proficient or above in math, up from 33 percent in 2003. In eighth-grade math 35 percent of students met proficiency, up from 32 percent four years ago. Results are similar in reading, with 35 percent of Nebraska's fourth and eighth graders meeting the proficiency bar.
Contrast these results with the proficiency rates on the state's annual academic report. In 2006-2007, Nebraska reported that 89 percent of fourth graders and 90 percent of eighth graders were proficient in reading and that 91 percent of fourth graders and 88 percent of eighth graders met the state's math-proficiency bar. These results do not jibe with the state's NAEP performance. To be fair, NAEP is a reality check for the rigor of Ohio's statewide assessments, too (see here). However, Ohio has seen larger gains than Nebraska in NAEP performance in recent years, especially in math.
There also are policy and implementation concerns related to decentralized assessments. Creating a fair, reliable, standards-based assessment at the state level has been a mammoth task. Imagine if that work were being done 612 times over by every district in the state. How would Ohio ensure that there were not vast differences between the expectations being set for urban children and suburban children? How would we accurately gauge whether achievement gaps exist and measure our progress toward closing them?
Ohio's students are mobile. In urban districts 15 percent of students tested on state assessments last spring were not in the same building for the entire school year. How would local assessments, unique to each district, measure the progress of students between districts? How would portfolios follow students across schools? And, to the bottom line, how much money and staff time would it take to implement such a system? Would that time and money be drained away from classroom instruction and resources?
Teachers should regularly review and monitor students' academic progress and tools like portfolios and short-cycle assessments are critical for this work. Classroom instruction should not be based solely on the content of annual statewide tests and local school districts should be encouraged to expand academic offerings beyond the state's standards. But this does not mean we ought to abandon statewide standardized testing. Accountability in education requires, at the very least, objective assessments based on common standards. Standardized testing, while not perfect, provides an important floor for public education and removing it would cost the state's neediest children most.
Ohio should continue thinking about other ways to measure school and student performance in addition to the statewide achievement tests, and we can learn from the strengths of systems like STARS. But we are making progress in Ohio and improving our accountability system with the new value-added component (see here). Doing away with statewide standardized testing would be a giant step back for children in the Buckeye State.