last updated: Jun 10 2007
Web Site Questions and Comments
The University of Texas at Austin
Executive Vice President and Provost

GSI
TA Talk Highlight

Improving Effectiveness and Efficiency in Grading
by Angela Bush, GRA, Center for Teaching Excellence

Accompanying the concerns of teaching course content effectively and creating assignments that will assess student learning are the related issues of how to grade students' understanding and evaluate their performance. As TAs and AIs we will or have already experienced the situation of sitting down with a stack of assignments and feeling excited to read student responses, yet sometimes a bit overwhelmed with all the considerations involved in grading. How can I be fair? How can I maintain consistency in grading over all assignments? How can I be sure that the grades assigned are effectively communicating to the student how they are doing? How can I establish a way of grading that is efficient in terms of the type and importance of the task? Will I grade based on whether students meet certain criteria or based on how an individual performs in relation to others in the class or a combination of both? If you have ever found yourself in a similar situation and want to further clarify how to address such issues, the following explanations of effective and efficient grading may be helpful in setting your mind at ease.

Dr. Marilla Svinicki, former Director for the Center for Teaching Excellence and Associate Professor in the Department of Educational Psychology, has enlightened professors, TAs, and AIs across campus and beyond for over 30 years. Her insights and achievements in higher education research and implementation are astounding, as was apparent when she delivered an outstanding discussion on the topic of effective and efficient grading at the 3rd Annual Seminar for Experienced TAs and AIs in January, 2003.

Grading Effectively

Grading should be supportive of student learning and accurately reflect their performance. To become a more effective grader, one should focus on what the student is learning and assisting them in knowing what to do about it. One way to grasp what it is you are grading for is by doing a critical components analysis. This involves identifying concrete ingredients that should be apparent in the assignment. Dr. Marilla Svinicki asked the TAs and AIs in the session to use a writing assignment to brainstorm components as an example. Identifying a clear thesis, showing evidence in support of the thesis, structuring an elegant argument, using appropriate language and vocabulary, and writing grammatically correct were discussed as benchmarks for grading such an assignment.

Next, to be effective in grading the components of an assignment, one should consider levels of performance for the critical components. Characteristics such as quality of the product, original thinking, and completeness in a timely fashion were decided as ways to weight levels of performance. Levels of the argument such as superior performance of original and clearly stated thesis; persuasive, well-organized, and imaginative use of source material to the minimum of no awareness of argument or complexity could be used to differentiate performance. Using descriptive scales of performance for each critical component increases the accuracy and reliability of grading. Establishing a rubric for grading does just that because we define the critical components and create a weighting system that is consistent within and across assignments. Not only should this be given back with the graded assignment, but allowing the students to have this before the assignment guides their efforts in applying it to their own work.

The above analysis of assignments using a grading rubric and weighting system is designed for grading using a criterion-referenced approach. It implies that one is evaluating on the basis of whether or not students meet certain standards. It is our hope that if we tell students what is expected of them from the beginning, they will rise to the challenge. Another approach is norm-referenced grading by which students are evaluated on the basis of the assumption that there is a normal distribution of performance in the class. Students' grades are assigned based on their performance in relation to other students in the class. This is most commonly known as grading on a curve. This approach to grading is not suggested for small classes and one should consider its potential to increase competition within a group of students.

Grading Efficiently

Effectiveness is certainly important to grading in all situations, but efficiency goes hand in hand in assessing student learning. "Am I grading accurately and consistently?" answers the question of whether the grading is reliable. Are the students receiving feedback in a timely way so that they can make use for further knowing what to do to improve is another important quality to consider in grading. Does the grading provide informational value as to why the grade assigned was the grade earned? The diagnostic component of grading should communicate what to do about it now that the assignment is completed and evaluated. To ensure the grading fits with the task, the grader should take into account whether the assignment is for a grade (summative assessment) or to help the students learn (formative assessment). If the purpose is for a grade, reliable and informative qualities of grading are crucial. If the purpose is leaning more towards student learning, the informative and diagnostic value of the grading as well as timeliness are key. It is imperative that students are able to use the information provided to know what to do next and how to go about successfully achieving that.

A matrix of common assignments and assessments to help guide efficient grading follows:

Further suggestions for making grading efficient include:

• Use comment summaries and let the students pick out their own or others' mistakes.
• Concentrate where they need the most help.
• Use only as many levels of grades as you need.
• Have students do some of the work by how they present the paper.
• Use technology to save time and enhance results.

Understanding and utilizing these ideas about how to make grading more manageable and explanatory to students will not only lessen the burden for the grader, but hopefully reduce tension between students and teaching assistants that is commonly associated with evaluating student performance. Consistency and accuracy should allow for mutual recognition by students, TAs, AIs, and professors concerning fair and reliable grading.