Students in Grades 5 and 8 completed a 30-minute writing performance assessment and a writing performance assessment completed over 3 days. Assessments were evaluated on four traits (ideas, organization, conventions, and sentence fluency). A significant interaction was found at Grade 5 between length of time allotted for the assessment and students' educational classification (general or special education). Grade 5 students performed significantly better on the 3-day writing assessment, with students in special education benefiting the most. At Grade 8, there were no differences between scores on the 30-minute and the 3-day assessments. No significant differences were found in students' writing performance across various types of discourse (narrative, imaginative, persuasive, and expository); significant differences were reported across certain writing traits. We shared three hypotheses for reported differences: (a) students' proficiency with the writing process, (b) scoring scale bias, and (c) differences in the revision behaviors of developing and mature writers. Educational implications related to statewide test programs are discussed.