Measuring Inconsistency in Written Feedback: A Case Study in Politeness

Feedback, indisputably, has been widely recognized as one of the most important forms of communication between teachers and students and a significant lever to enhance learning experience and success. However, there is consistent evidence showing that higher education institutions struggle to deliver consistent, timely, and constructive feedback to students. This study aimed to investigate whether, and to what extent, feedback inconsistency manifested itself in terms of politeness displayed to students of different demographic attributes (i.e., gender and first-language background). To this end, a large-scale dataset consisting of longitudinal feedback given to 3,249 higher-education students in 35 courses were collected and analyzed by applying multi-level regression modeling. We demonstrated that there were significant differences between low-performing and high-performing students as well as between English-as-second-language and English-as-first-language students. However, the majority of variance measured in the politeness of feedback was explained by course-level and assessment-level characteristics, while student-level characteristics accounted for less than 1% variance.