“Learning Analytics is the measurement, collection, analysis and reporting of data about learners and their contexts, for purposes of understanding and optimizing learning and the environments in which it occurs.” — from SOLAR (Society for Learning Analytics Research)
There are several reading materials recommended here, they help frame this topic, and define terms and concepts.
(from EDUCAUSE Review)
A clear distinction is made between learning analytics and academic analytics:
- Learning analytics is concerned with understanding and optimising learning and the environments in which it occurs, and operates at course and departmental level.
- Academic analytics applies business intelligence in an education context and operates at institutional, regional, national and international level.
Within the learning analytics domain, Siemens suggests a hierarchy framework to help position your learning analytics initiative.
- Course-level: this is the starting level in the hierarchy and looks at learning trails, social network analysis and discourse analysis, this is learning analytics at its most basic and an example would be a learning management system analytics tool.
- Educational data-mining: This is the next level and is concerned with predictive modeling, clustering and pattern mining, focusing on identifying patterns of success at departmental level.
- Intelligent curriculum: this is the development of curricular resources mapped against a specific knowledge domain. Assessment is performed almost constantly and content is provided on the fly in response to the learner’s profile, learning goals and the knowledge domain.
- Adaptive content: this takes things a step further and aims to deliver an adaptive sequence of content based on learner behavior and recommender systems. Applications such as Knewton operate in this space.
- Adaptive learning: so I guess this is the holy grail of learning analytics, and it relates to an adaptive learning process, which includes adapting social interactions, learning activity and learner support, not just content.
(from Wolfgang Greller and Hendrik Drachsler, Open University of the Netherlands)
Six critical dimensions (soft and hard) of Learning Analytics(LA) need to covered by the design to ensure an appropriate exploitation of LA in an educationally beneficial way. The author aim at a description framework that can later be developed into a domain model or ontology. The dimensions are considered “critical” in the sense that each of the six fields of attention is required to have at least one instantiation present in a fully formulated LA design. Each of these dimensions is discussed individually in the paper.
Created out of a number of instantiations of the six dimensions, a specific example related to conducting a social network analysis of students discussing in a forum using the SNAPP tool, based on the work by Dawson et al. (Dawson, 2008; Macfadyen & Dawson, 2010) is presented.
Learning Analytics Workflow Model
Siemens, G. (2013). Learning Analytics: The Emergence of a Discipline. American Behavioral Scientist, (August).
Collection & Acquisition: The extraction of source data
Storage: Storing source data in a data warehouse, where relevant
Cleaning: Rectifying anomalies and inconsistencies, and normalising the syntax of the data
Integration: Aligning the data to either existing datasets, or a common vocabulary
Analysis: Analyse the data, in order to build descriptive or predictive models
Representation and Visualisation: Creating reports and diagrams that illustrate the models for a wider audience
Alerting: Operationalising the models to (near) real time to enable alerting of relevant stakeholders