Supporting TA Training and Management Using an Evidence-Based Approach: A Visual Tool to Identify Outliers in Student Grades

 

Tested Studies in Laboratory Teaching, 2016, Volume 37

Liane Chen, Megan Barker, & Laura Weir

Abstract

In large classes with multiple lab/tutorial sections and graduate teaching assistants (TAs), it can be a challenge to determine whether grade variations between TAs are a result of differences in grading or whether they reflect other factors such as underlying student ability. Instructors can assess grading quality by checking a subset of graded work, but the number of TAs and the amount of graded work to check can be prohibitive. Gillian Gass and I previously developed an Excel-based graphing tool that created a visual depiction of the grade distribution within and between TA groups, allowing instructors to easily identify outliers (Proceedings of the Association for Biology Laboratory Education 34:310-313). Subsequent work has demonstrated that most outliers identified visually are also detected through statistical tests, underlying the rigor of the tool despite its simplicity of use. Here, participants are invited to interpret typical graphs and analyze their own class data, in order to propose underlying causes of grade outliers and to discuss subsequent TA training interventions.

Keywords:  teaching assistants, grading, graduate teaching assistants, variations in grading

Boston University (2015)