Community colleges now have the opportunity to benefit from Big Data by identifying predictive indicators for successful and unsuccessful students in their pursuit of a college education. The term Big Data refers to the use of large data sets that would be unrealistic for any one person to analyze manually (Ferguson, 2012). Making use of large amounts of data, including pre-existing learning management system information, behavior and service use (or lack thereof) are criteria that can now be analyzed to make predictions on the probability of students’ success. Historically, in higher education these large data sets have not been considered to predict user behavior, since predictive analytics has primarily resided in businesses such as Amazon and Google for use in product matching and contextual advertising.
Problems these large data sets would solve in community colleges include learning more about students and student behaviors. With more students opting to take courses online, the use of data can help predict needed supports and opportunities in online learning through the use of course-level and student-level data (Ferguson, 2012). Additionally, with Big Data and data visualization software, faculty and administrators can have accessible data that assist in course development, pedagogy, and service delivery.
As institutions prepare to roll out the use of Big Data, attention should be placed on ensuring all stakeholders are involved with the rollout to the greater campus community. Without all parties being involved, faculty may be concerned with the use of Big Data, issues of privacy, or the possibility for institutions to use Big Data for program cuts or cut backs to positions (Gagliardi & Wilkinson, 2017).
The use of Big Data can help support department planning, resource allocation, accreditation, and scheduling. Colleges and universities have used Educational Data Mining to engage with students in a way that enhances student success. For example, Purdue University created and implemented the Signals initiative, where faculty notified students when they were on track, off track, or getting close to being off track by assigning a green, red, or yellow colored light on the student portal early in the term. Results from Signal are promising with gains in student retention, persistence, and completion (Pistilli, Arnold, & Bethune, 2013).
To help implement the use of Big Data, institutions will need to have a clearly articulated data governance structure that provides the same opportunity for everyone to access data in the same fashion. Removing the stigma that data belongs to the Institutional Research office will be key to the sustainability of campus-wide adoption of Big Data use.
References
Ferguson, R. (2012). Learning analytics: drivers, developments and challenges. International Journal of Technology Enhanced Learning, 4(5/6), 304-317. Retrieved from http://oro.open.ac.uk/36374/1/IJTEL40501_Ferguson%2520Jan%25202013.pdf
Gagliardi, J. S. and Wilkinson, P. (2017, December). How colleges and universities can use analytical resources to better serve their missions. Retrieved from https://www.higheredtoday.org/2017/12/13/big-data-campus/
Pistilli, M., Arnold, K., & Bethune, M. (2013, July). Signals: Using academic analytics to promote student success. Retrieved from http://www.educause.edu/ero/article/signals-using-academic-analytics-promote-student-success
