Article originally published on E-Learn Magazine on Jan 23, 2018
Learning analytics aims to provide students, instructors, and institutional leaders with relevant information that can help them distill and optimize the teaching and learning experience. Although several institutions have matured quickly in their use of educational data, the field of learning analytics is still very young, and many colleges and universities are in the very early stages of the adoption process. The following definitions are designed to provide a helpful starting point into this field of research and practice.
Adaptive Learning - Adaptive learning usually involves software that observes student performance and adjusts what it presents to each student based on those observations.1
Artificial Intelligence – A branch of computer science that aims to build machines capable of simulating the human decision-making process.2 In practice, artificial intelligence can vary widely in sophistication, at times employing hard-coded ‘triggers,’ and at other times relying on complex machine learning algorithms. Educational institutions explore the use of artificial intelligence through faculty and student data collection to understand, model, predict and automate processes to improve teaching, learning, and student success.3
Big Data - Extremely large or complex data sets that may be analyzed computationally to reveal patterns, trends, and associations which cannot be processed by traditional data processing application software. Higher education, in particular, can generate rich data sets through Big Data. The challenge is to distill the data into useful information for the benefit of students, instructors, and institutions.
Code of Practice – In learning analytics, it is the written guidelines that set out the responsibilities of educational institutions to ensure that learning analytics is carried out ethically, responsibly, appropriately, and effectively.4
Cognitive Computing – A combination of cognitive science — the study of the human brain and how it functions — and computer science. The goal of cognitive computing is to simulate human thought processes through a computerized model. The computer can mimic the way the human brain works by using self-learning algorithms that use data mining, pattern recognition, and natural language processing.5
Dashboard – A collection of widgets that provide the user with an overview of the reports and metrics needed to achieve one or more objectives. In learning analytics, they’re primarily intended for faculty, administrators, and other professionals, but students can also benefit from specific dashboards.6
Data Democratization – The ability for information in digital formats to be accessible to everyone in a given organization, not just specialists or managers, so they may gather and analyze data by themselves. Data warehouse solutions can support data democratization for educational institutions, enabling administrators, instructional designers, instructors, and students to align what happens in the classroom with the institution’s retention or graduation goals.
Data Governance – The management of policies, systems, security, and practices, in order to ensure that an institution’s data is accurate, complete, consistent, reliable, and available to the right people at the right time.
Data Science – A science that combines different statistical, computational and visualization methods in order to derive meaningful information from large data sets. Data Science is often applied to create predictions that other processes rely on, or to understand complex phenomena. Aside from education, it is also applied in fields such as finance, sports, biological sciences, public health, astronomy, and internet activity.
Data Warehouse – The electronic storage of corporate or institutional data. With a data warehouse, data is validated and systematically checked so that inconsistencies are eliminated and standard data definitions are preserved. A centralized, trustworthy, authoritative, and accessible source of institutional information improves communication and supports decision-making.7
Discourse Analytics – Within the field of learning analytics, it typically includes processing discussions occurring in a virtual learning environment, such as discussion forums, chat rooms, blogs, and even wikis. It aims to capture meaningful data on student interactions to explore the properties of the language used.
Educational Data Mining (EDM) – A discipline focused on the development of methods for exploring the data that comes from educational settings and using those methods to better understand students, and the settings in which they learn.8 It is closely related to learning analytics, which places more emphasis on simultaneously investigating automatically collected data, along with human observation of the teaching and learning context.9 The main goal of both EDM and learning analytics is to extract information from educational data to support education-related decision-making. From a general perspective, EDM focuses more on techniques and methodologies, while learning analytics deals more with applications.10
Internet of Things – A concept comprising the everyday objects that are connected to the internet and are able to collect and exchange data, relating to other devices and databases. Internet-connected objects contain sensors that can be used to collect data, and which increase the visibility of researchers into the teaching and learning process. Some examples of Internet of Things applications in education are mobile learning, smart lighting, and security systems on campuses and smart boards.
Learning Analytics – Learning analytics uses data about students and their learning to help understand and improve educational processes, and to assist the learners themselves.11 It covers a range of methods that include machine learning, dashboard design, social network analysis, writing analytics, and natural language processing.
Machine Learning – The science of getting computers to learn and act like humans do and improve their learning over time in an autonomous fashion, by feeding them data and information in the form of observations and real-world interaction.12 Predictive analytics is an example of machine learning application in education.
Nudging – It involves structuring an environment in such a way as to encourage a small set of behaviors, without also actively limiting an individual’s ability to freely choose from a much wider range of options. In learning analytics, for example, it may mean providing students with information about how they are performing and alerting them in advance if they are at risk of not passing a course. It may also mean using activity data to understand and scale instructional design patterns that are likely to improve student course engagement.
Predictive Analytics – Predictive analytics can help instructors understand the probability of future events occurring by analyzing historical and current data and answering questions such as: “Why is this happening?”, “What if these trends continue?”, “What will happen next?” and “What is the best that can happen?”13 In education, predictive models can use years of student demographic and performance data to generate forecasts about which students are likely to struggle. It helps in shaping positive outcomes related to student success while there is still time to act.14
Self-regulated Learning – A sense of personal responsibility for one’s learning. When made available to students, learning analytics can provide them with the information needed to help raise awareness about their own learning. For example, benchmarking student activity against their peers can promote reflection about the relationship between performance and relative effort, and encourage students to adopt behaviors that are more likely to produce the desired result.
Social Network Analysis (SNA) – A procedure that allows the study of interactions and the strength of relations between individuals in a social network. In the field of learning analytics, social network analysis can help instructors and instructional designers to assess student engagement in discussion boards, as well as the impact that specific assignments may have on a network’s shape. Using SNA to identify weak ties can also help identify students at risk of dropping out. When combined with writing analytics and natural language processing, SNA can help instructors understand individual learners’ discussion forum activity, as well as critical thinking and originality of contributions.
Writing Analytics – The measurement and analysis of texts written by students to understand writing processes and products in their own contexts. It aims to employ learning analytics to develop a deeper understanding of writing skills.15
1 Feldstein, M. What Faculty Should Know About Adaptive Learning. Retrieved November 14, 2017, from http://mfeldstein.com/faculty-know-adaptive-learning/
2 Bell, L. (2016, December 1). Machine learning versus AI: what’s the difference? Retrieved November 14, 2017, from http://www.wired.co.uk/article/machine-learning-ai-explained
3 Davis, V. (2017, November 1). The Ethical and Legal Dimensions of AI. Retrieved November 14, 2017, from http://blog.blackboard.com/the-ethical-and-legal-dimensions-of-ai/
4 Sclater, N., & Bailey, P. (n.d.). Code of practice for learning analytics. Retrieved November 14, 2017, from https://www.jisc.ac.uk/guides/code-of-practice-for-learning-analytics
5 Marr, B. (2016, March 23). What Everyone Should Know About Cognitive Computing. Retrieved November 14, 2017, from https://www.forbes.com/sites/bernardmarr/2016/03/23/what-everyone-should-know-about-cognitive-computing/#3b98f3ab5088
6 Whitmer, John. (2017, February 2). Surprising lessons from research on student feedback about data dashboards. Retrieved November 14, 2017, from http://blog.blackboard.com/research-student-feedback-data-dashboards/
7 Blackboard. (n.d.). Blackboard Intelligence. Retrieved November 14, 2017, from http://www.blackboard.com/resources/pdf/datasheet-blackboardintelligence-rev20170209.pdf
8 Educational Data Mining. (n.d.). Educational Data Mining. Retrieved November 14, 2017, from http://educationaldatamining.org/
9 The Center for Innovative Research in CyberLearning. (n.d.). Educational Data Mining and Learning Analytics. Retrieved November 14, 2017, from http://circlcenter.org/educational-data-mining-learning-analytics/
10 Liñán, Laura and Pérez, Angel. Educational Data Mining and Learning Analytics: differences, similarities, and time evolution. Retrieved November 14, 2017, from http://rusc.uoc.edu/rusc/ca/index.php/rusc/article/view/v12n3-calvet-juan/2746.html
11 SCLATER, Niall. (2016, September 01). What is learning analytics and how can it help your institution? Retrieved November 13, 2017, from http://sclater.com/blog/what-is-learning-analytics-and-how-can-it-help-your-institution/
12 Faggella, Daniel. What is Machine Learning? Retrieved December 2, 2017, from https://www.techemergence.com/what-is-machine-learning/
13 B. D., & J. H. (n.d.). Using Learning Analytics to Predict (and Improve) Student Success: A Faculty Perspective. Retrieved November 14, 2017, from http://www.ncolr.org/jiol/issues/pdf/12.1.2.pdf
14 Rattiner, M. (2017, May 19). Walking the line of predictive analytics in higher education. Retrieved November 14, 2017, from http://blog.blackboard.com/walking-the-line-predictive-analytics-higher-education/
15 Writing analytics (n.d). LAK16: Critical Perspectives on Writing Analytics. Retrieved November 14, 2017, from http://wa.utscic.edu.au/events/lak16wa/