Concerns about the instructional validity of traditional assessment have led researchers and teachers to explore different ways to evaluate students' performance on general literacy abilities. Valid assessment has particular significance for children with special needs, for whom effective instruction is so closely tied to accurate and appropriate diagnosis. Among the most commonly used forms of alternative assessment are literacy portfolios. While portfolio assessment appears to be a promising practice, teachers, administrators, and parents have raised questions about two particular aspects: (1) establishing shared standards across classrooms and grades and (2) achieving agreement in judging individual pieces of work. The purpose of this article is to describe one framework for using portfolios in which common benchmarks and rubrics provided explicit and shared criteria for judging both the collection of work in the portfolio and individual performance samples.