Content area
Full Text
Abstract
The educational system involves a complex set of actors, including learners, parents, teachers, and administrators. However, we now have more data than ever to analyze this system, which could result in a quick understanding and evaluation of public policies in this complex policy area. This paper explores a new area of data about the educational experience, namely social media data. This paper outlines an exploratory analysis of the Twitter discussions regarding higher education in the USA. Based on a collection of more than 1.5 million tweets over a period of 4 months, we identify a few key issues in the current higher education discourse on social media. We also identify the effect of the expressed feelings of the social media users when it comes to college applications, decisions and completion. We conclude that policies in higher education can be better tailored if they are informed by social media discussions.
Introduction
THE INCREASING AMOUNT OF DATA, the decreasing cost of computational power, and the improving state of analytics has revolutionized fields from stock trading to social analytics, but somehow higher education has not received as much attention. The technology that has transformed many forprofit businesses and governments can be applied at various colleges and universities.
One obvious place that analytics could be useful is in the classroom, but currently instructors at many universities are using outdated and inefficient methods to grade assignments and compile these scores into selfgenerated databases. In fact, Darrell West argues that "many of the typical pedagogies provide little immediate feedback to students, require teachers to spend hours grading routine assignments, are not very proactive about showing students how to improve comprehension, and fail to take advantage of digital resources that can improve the learning process" (West 2012). Data mining and analytics provide the capabilities necessary to circumvent the traditionally cumbersome grading processes and glean insights from student data about performance, learning approaches, and other metrics. For example, Leah Macfadyen and Shane Dawson developed an "early warning system" which correctly identified 81% of students who failed an online course by creating a regression model that analyzed such variables as total number of discussion messages posted and total number of assignments completed (Macfadyen and Dawson 2010).
Big data analytics within education...