Learning Analytics

Ari Korhonen, Jari Multisilta

Research output: Chapter in Book/Report/Conference proceedingChapterScientificpeer-review

Abstract

In online learning, many Learning Management Systems (LMS) and Massive Online Open Courses (MOOC) collect data from their users. The field studying the use of the educational data is known as Educational Data Mining (EDM) or Learning Analytics (LA). One of the main reasons for creating and giving open online courses is the rich data which is possible to collect from the learning process and students. This data can include background information about the students, forum discussions, log data collected from the system usage, the solutions to the problems submitted by the students, and much more. Especially in scalable courses that are targeted to large number of students, automatic assessment and feedback provides an interesting source of data. The learning platform can collect not only the submitted solutions and the corresponding grading information (such as points or other marks awarded to the student), but it also traces the paths a student took while solving the problems. General analytics tools, such as Google Analytics, can be used to evaluate the use of a learning service, but it provides only very superficial information on the learning process. Generic data includes web navigation data such as page hits, the number of visitors to a page and time spent on the page. We claim that there is a great need to utilize learning related user data to support the understanding of the learning process in much deeper way. To be able to do so, learning data should be collected and analyzed in relation to the content; and the content needs to be structured in a pedagogically meaningful way. The many data sources allow us to ask research questions from multiple perspectives. Real time data analysis can provide instant feedback loops for the learners and instructors, e.g., in the form of progress reports. The same data can be utilized to develop the system further in the long run. Likewise, institutions are nowadays interested in monitoring courses and sharing best practices. The main goal is to develop courses based on data analysis. This is achieved by organizing experimental studies (A/B testing), applying (educational) data mining techniques, and employing machine learning methods to make predictions on data. In this paper, we discuss the future of learning analytics and evidencebased development of online courses.
Original languageEnglish
Title of host publicationNew Ways to Teach and Learn in China and Finland
Subtitle of host publicationCrossing Boundaries with Technology
EditorsHannele Niemi, Jiyou Jia
Place of PublicationFrankfurt am Main
PublisherPeter Lang Verlag
Pages301-310
Number of pages10
ISBN (Electronic)978-3-631-69874-7
ISBN (Print)978-3-631-67642-4
DOIs
Publication statusPublished - 2016
MoE publication typeA3 Part of a book or another research book

Fingerprint Dive into the research topics of 'Learning Analytics'. Together they form a unique fingerprint.

  • Cite this

    Korhonen, A., & Multisilta, J. (2016). Learning Analytics. In H. Niemi, & J. Jia (Eds.), New Ways to Teach and Learn in China and Finland: Crossing Boundaries with Technology (pp. 301-310). Frankfurt am Main : Peter Lang Verlag. https://doi.org/10.3726/978-3-631-69873-0