By John R. Hayes, Jill A. Hatch, Christine Murphy Silk. Quarterly of the National Writing Project and the Center for the Study of Writing and Literacy, v. 17 no. 4 p. 34-36, Fall 1995.
Abstract: If a student writes a B+ essay at the beginning of the semester, is it reasonable for the instructor to expect that the student will consistently write above-average essays throughout the semester? In this study, we asked independent, experienced writing instructors to judge a series of writing samples without knowing the nature of the study, the students, or whether the writing samples were even written by the same group of students. Our results show that essays written by a particular student did not receive consistent grades from a given judge (correlations averaged less than .2). One important implication of our findings is that just because a student performs well on one particular writing task does not tell us how well he or she will perform on subsequent writing tasks. Click here for the full article.