Supporting students’ peer assessment using chatbots

Peer assessment is found to be an effective approach for instructors to evaluate students’ learning performance in large classes, and it provides great learning opportunities for stu- dents. However, conducting successful peer assessment remains challenging as peer graders often do not have sufficient guidance to make fair assessment and to provide useful feedback. In this thesis, we explore the potential of using a chatbot to facilitate the execution of peer assessment. We conducted two studies in two university courses. The first one was a within- subject study, which focused on understanding the effect of chatbot use for peer assessment over time. The second was a between-subject study, which addressed the differences between providing guidance by a chatbot and giving suggestions through a regular form. Our results show that, using a chatbot for several times, students improved the performance of conducting peer assessment by providing more accurate scores that were closer to the TAs’ scores. Also, when the chatbot provided guidance, high-performing peer graders were more likely to give positive comments with constructive suggestions than low-performing peer graders did. Our results provide insights into designing interaction technologies to improve both the quality of peer assessment and students’ learning.

Name thesis:degree_name

M.S.

Level thesis:degree_level

Thesis

Discipline thesis:degree_discipline

Computer Science

Grantor

University of Illinois at Urbana-Champaign

Year dc:date

2020

Author dc:creator

• Wang, Dennis

Contributors dc:contributor

• Huang, Yun

• Peer Assessment
• Chatbot
• Conversational Agent

Statement dc:rights

• Copyright 2020 Dennis Wang

Language dc:language

en

Handle dc:identifier

http://hdl.handle.net/2142/108353

OAI record oai:identifier

https://www.ideals.illinois.edu/oai-pmh?verb=GetRecord&metadataPrefix=etdms&identifier=oai%3Awww.ideals.illinois.edu%3A2142%2F108353