A common characteristic of a standards-based curriculum is the presence of performance-based tasks designed to assess students' knowledge of skills and performance. Much of the subjectivity inherent in such assignments can be minimized by creating rubrics according to predetermined expectations and criteria (Taggert, et.al, 1998). The purpose of this study was to have pre-service teachers (n=16) assess fifth-graders' written assignments initially without rubrics and a second time with rubrics. Fifth grade students in health and physical education class were taught a lesson about the brain. The students' assignment was to explain and illustrate four ways to have a healthy brain using information from the health lesson. Each item on the assessment included a need (e.g., energy), solution (e.g., carbohydrates), example (e.g., rice) and amount (e.g., 6 servings). Four of the student papers were selected for use in this study based on their level of accuracy. The first paper had all the criteria correct, the second paper had at least 1 or 2 inaccuracies across the 4 items, the third paper had more than three inaccuracies across each item, and the fourth paper was mixed with one item correct the other 3 items with different levels of inaccuracies. Subjects initially assessed the papers without using rubrics, and a second time using rubrics (scoring range=5-20). They understood the rubrics and their application from the written and oral instructions given by the investigator. When all 4 parts of an item were explained correctly in words or pictures, a score of 5 was given. When 2 parts of an item (need and solution) were represented correctly a score of 3 was given. When one part of an item (need or solution) was explained correctly a score of 1 was given. A comparison of scores using a repeated measures ANOVA indicated significant differences between the pre- service teachers' scores with and without the use of rubrics. Regardless of the errors in a paper, all scores dropped significantly (p<.05) when using rubrics with a range of .06 - 6.38 on a 20 point scale. Larger drops between scores were associated with papers that had more errors. Based on the use of rubrics, the researchers concluded there was a tendency for teachers to significantly inflate the scores when no objective scoring process was in place. The results confirm the need for such a process.