Using an AI-Supported Online Discussion Forum to Deepen Learning

Introduction

Online discussions provide a rich environment for student learning as they serve to facilitate peer-to-peer interactions and knowledge transfer. Rich discussions in face-to-face settings are common. When implemented thoughtfully, online discussion tools can serve as a way to ensure that these learning opportunities are available in fully online classes as well. While there are many benefits to using online discussion forums, such as opportunities for regular engagement with coursework (Pazzalgia, Clements, Lavigne, & Stafford, 2016), the ability to review discussions after class (Kinskey, Miller, Hauck, & Manderfield, 2018), allowing time for reflection before posting, and the ability for “instant replay” (Cox, 2011), several challenges also exist. For example, it can be difficult to encourage deep and meaningful interaction in online classes where students may lack the motivation to participate in what may feel like “busy work” (Cox, 2011). Students may also feel disconnected from their peers and therefore be less interested in what they might have to say. In addition, assessing these activities, particularly in large-enrollment classes, can be time consuming and, at times, feel almost impossible. Often, for students to feel motivated to participate in online discussions, clear expectations need to be communicated regarding how the discussion will tie into their grade; however, even more difficult than setting expectations is building in incentives for active participation in learning through online discussions (Cox, 2011). Discussion forums are often the “heart” of an online course (Cox, 2011). Consequently, it is essential that instructors put in the thought and work required to make these discussions meaningful. Arend (2009) found that “the way discussions are used and facilitated is vital for encouraging critical thinking.”

Two professors at Texas State University sought to change their approach to discussion forums in their online courses to address issues of engagement and meaningful evaluation, which they found to be lacking when using traditional learning management system (LMS) forum tools. Specifically, they individually and independently elected to use a third-party online discussion tool called Packback that uses algorithms derived from artificial intelligence (AI) to determine the quality of student posts. According to the Packback website (https://packback.co, 2019), the company has 247,000 students from across 208 institutions and 1,323 instructors posting over 5 million questions and responses since May 2016. The creators at Packback describe their tool as:

After the pilot implementation by these two professors, a look at how the tool may have improved some areas of identified concern in the literature was warranted. The purpose of this study was to determine if the use of AI-supported platforms for online discussions helped address common particularly challenging issues including student motivation, development of students’ critical thinking, and ease of instructor assessment. This study also aims to ascertain the value of Packback, particularly with regards to its efficacy in enhancing student engagement and outcomes in online discussions.

Background

Articles regarding the impact of online discussion forums on students began appearing in the literature around the turn of the century. These articles are replete with suggestions for instructor presence, management, and student motivation ideas, and mostly measure student satisfaction, which Allen, Bourhis, Burell & Mabry (2002) understood is impacted directly by teacher-student and student-student interactions. Without a doubt, the literature identifies number of challenges in the use of online discussion forums.