Throw 'em out or make 'em better? State and district high-stakes writing assessments

The writing of school-aged children is assessed for many reasons (Graham, Harris, & Hebert, 2011). Teachers assess writing to monitor students' growth as writers, inform instruction, provide feedback, and evaluate the effectiveness of their teaching. Students assess their own writing to appraise growth, identify strengths, and determine areas in need of further development. Peers assess other students' writing to provide them with feedback on what works in a paper and what still needs work. States and school districts assess writing to determine how many students meet local or state performance standards, identify youngsters who need extra help, and evaluate the effectiveness of individual teachers and schools. The national government administers the National Assessment of Educational Progress (NAEP) writing test to measure American students' collective writing success, evaluating students' writing performance across time. Given the heavy emphasis now placed on assessment and evaluation as a tool for improving and reforming writing and other aspects of education in the United States (Gewertz, & Robelen, 2010; National Commission on Writing, 2003), it is important to ask whether the various forms of assessment, ranging from classroom-based writing assessments to state and district evaluations (the focus of this article) do, in fact, make a difference in improving how well students write? For students with disabilities, such questions are especially important, as so many of these students experience difficulty learning to write. On the 2007 NAEP (Salahu-Din, Persky, & Miller, 2008), just 6% of eighth-grade and 5% of twelfth-grade students with disabilities performed at or above the "proficient" level in writing (defined as solid academic performance). Students scoring below this level are classified as obtaining only partial mastery of the literacy skills needed at their respective grade. Thus, this assessment indicates that 19 of every 20 students with disabilities do not acquire the writing skills needed for success in school. Although classroom-based assessments are not the focus of this article, evidence shows that such assessments can make a difference in improving how well students write. A recent meta-analysis of experimental and quasi-experimental studies conducted mostly with typically developing students (Graham, Kiuhara, McKeown, & Harris, 2011 ) provided empirical evidence that writing assessments that are part of typical classroom practices improve the overall quality of students' writing. When students receive feedback about their writing and learning progress, writing improves. When students evaluate their own writing, writing