Improving Online Course Performance Through Customization: An Empirical Study Using Business Analytics

Improving Online Course Performance Through Customization: An Empirical Study Using Business Analytics

Siva Sankaran, Kris Sankaran
Copyright: © 2016 |Pages: 20
DOI: 10.4018/IJBAN.2016100101
OnDemand:
(Individual Articles)
Available
$37.50
No Current Special Offers
TOTAL SAVINGS: $37.50

Abstract

The number of educational courses offered online is growing, with students often having no choice for alternative formats. However, personal characteristics may affect online academic performance. In this study, the authors apply two business analytics methods - multiple linear/polynomial regression and generalized additive modeling (GAM) - to predict online student performance based on six personal characteristics. These characteristics are: communication aptitude, desire to learn, escapism, hours studied, gender, and English as a Second Language. Survey data from 168 students were partitioned into training/validation sets and the best fit models from the training data were tested on the validation data. While the regression method outdid the GAM at predicting student performance overall, the GAM explained the performance behavior better over various predictor intervals using natural splines. The study confirms the usefulness of business analytics methods and presents implications for college administrators and faculty to optimize individual student online learning.
Article Preview
Top

Prior Research

Among factors that have been found to affect a student’s level of success are: course prerequisite preparation (Marcus, 2013), time management skills (iSeek, 2015), technology infrastructure (Kukulska-Hulme & Jones, 2012), course content, structure and delivery (Teng, 2008). Other factors include getting familiar with the technology platform and course set up early, maintaining regular communication with the instructor, committing to consistent online study sessions, initiating ties with coursemates, university policies and netiquette (Harrell & Bower, 2011; Harrell & McClinton, 2016). We can group the above factors broadly into two categories: 1) System factors, and 2) Student background characteristics.

System Factors

According to Moore and Kearsley (2012), an online education system involves technology, courses, students, instructors and administrators. Technology includes access and software, whereas courses cover content and pedagogy. Students need to have prerequisite characteristics such as communication abilities and motivation. Instructors are to be technologically proficient and committed, and administrators must provide resources (Tirziu & Vrabie, 2015). Currently, many online systems are asynchronous implying students are separated geographically and in time. Synchronous systems may allow video and voice, either pre-recorded or in real time, to simulate a classroom (Latchman, Akkaraju & Gharbaran, 2010). Such systems can also capture social presence very effectively (Kear, Chetwynd & Jefferis, 2014). Lo, Chan and Yeh (2012) developed a customized learning system using neural network and demonstrated it can adapt to student cognitive styles dynamically through prior interactive sessions. However, research in customization is ongoing (Guthrie, 2012).

Complete Article List

Search this Journal:
Reset
Volume 11: 1 Issue (2024)
Volume 10: 1 Issue (2023)
Volume 9: 6 Issues (2022): 4 Released, 2 Forthcoming
Volume 8: 4 Issues (2021)
Volume 7: 4 Issues (2020)
Volume 6: 4 Issues (2019)
Volume 5: 4 Issues (2018)
Volume 4: 4 Issues (2017)
Volume 3: 4 Issues (2016)
Volume 2: 4 Issues (2015)
Volume 1: 4 Issues (2014)
View Complete Journal Contents Listing