| The practice of grading student writing often seems arbitrary and unfair to students. It is important to students and educators that writing is fairly and accurately assessed in order to facilitate demonstrable improvement in their composition. This research addressed a specific writing rubric as a method to mitigate implicit or subconscious biases before it could affect the grading process. The study was grounded in critical race theory, which in part states that certain kinds of biases are normal, but can be interpreted negatively by those affected. This experimental design first tested for name bias. This research also compared a writing rubric against a simple grading scale to test for bias mitigation. In this 2X3 Analysis of Variance (ANOVA) design, 82 middle school teachers were given the same middle school-level essay to grade. Half were given a specific writing rubric to use while grading; the others were given only a simple grading scale. Two different student names were used, which would signal the gender and ethnicity of the student writer. A control group using each of 2 grading methods was given a student essay with no name. Rubric scores were analyzed using an ANOVA, with Tukey's test for Honest Significant Difference refining the data. A finding of name bias was not found in any of the analyses. Additionally, the writing rubric and scaled scoring methods showed no significant difference in scores among participant groups. These findings, though unexpected, have important implications for social change. They illustrate that grading is not an exact science, even when careful scientific assessment practices are used to try to improve it. Despite negative findings in this case, the study is grounded in current literature, theory, and research methods; as such, it can serve as a model to educators for finding and mitigating implicit biases that all teachers struggle to keep out of the grading process. |
