This report describes the development of grade norms for timed-writing performance in two modes of writing: persuasive and descriptive. These norms are based on objective and automatically computed measures of writing quality in grammar, usage, mechanics, style, vocabulary, organization, and development. These measures are also used in the automated essay scoring system "e-rater"® V.2. Norms were developed through a large-scale data collection effort that involved a national sample of 170 schools, more than 500 classes from 4th, 6th, 8th, 10th, and 12th grades and more than 12,000 students. Personal and school background information was also collected. These students wrote (in 30-minute sessions) up to 4 essays (2 in each mode of writing) on topics selected from a pool of 20 topics. The data allowed us to explore a range of questions about the development and nature of writing proficiency. Specifically, this paper provides a description of the trajectory of development in writing performance from 4th grade to 12th grade. The validity of a single developmental writing scale is examined through a human scoring experiment and a longitudinal study. The validity of the single scale is further explored through a factor analysis (exploratory and confirmatory) of the internal structure of writing performance and changes in this structure from 4th grade to 12th grade. The paper also explores important factors affecting performance, including prompt difficulty, writing mode, and student background (gender, ethnicity, and English language background).