As previously noted, holistic scoring gives students a single, overall assessment score for the paper as a whole. Although the scoring rubric for holistic scoring will lay out specific criteria just as the rubric for analytic scoring does, readers do not assign a score for each criterion in holistic scoring. Rather, as they read, they balance strengths and weaknesses among the various criteria to arrive at an overall assessment of success or effectiveness of a paper. The CSU composition placement exam (administered from 1977-2004 and then replaced by the Composition Challenge Exam for a smaller number of students) relied for many years on a 9-point scale for overall assessment. Although the composition program now uses a 6-point scale, the rubric functions in much the same way. Notice that the four key criteria are defined most concretely for "upper-range" papers. Deficits from the most effective demonstration of the criteria characterize the "mid-range" and "lower-range" papers.
A reader writes nothing on the paper itself and assigns the holistic score after reading the paper carefully and completely. A second reader, who does not see the first score, independently reads and assigns a second holistic score. If the two scores differ by more than 2 points, then a third reader scores the paper as well. Inter-rater reliability (the percentage of papers given the same score or differing by one point) should fall between .85 and .90 for sound holistic scoring. Readers who read the same kinds of papers regularly (including students in a large class) can easily be trained to reach acceptable inter-rater reliability scores.
AP exams and the SAT II writing test both use holistic scoring to assess student writing skills.