Learning Analytics

Learning Analytics

Luis de-la-Fuente-Valentín (Universidad Internacional de La Rioja, Spain), Alberto Corbi (Universidad Internacional de La Rioja, Spain), Rubén González Crespo (Universidad Internacional de La Rioja, Spain) and Daniel Burgos (Universidad Internacional de La Rioja, Spain)
DOI: 10.4018/978-1-4666-5888-2.ch231
OnDemand PDF Download:
$30.00
List Price: $37.50

Chapter Preview

Top

Background

Learning analytics is a relatively new field that has drawn techniques widely used in a number of communities. Some of them, as highlighted by Cooper (2012), are Statistics, Bussiness Intelligence, Web analytics or Operational research. The use of Analytics tecniques in the context of the learning process is what the community call Learning Analytics. A widely accepted definition of Learning Analytics, provided at the 2012 International Conference on Learning Analytics and Knowledge, describes the field as “the measurement, collection, analysis and reporting of data about learners and their contexts, for [the] purpose of understanding and optimizing learning and the environments in which it occurs” (Siemens, 2012). To discuss such definition, it’s necessary to focus on its three main parts, namely: procedures (measurement, collection, analysis and reporting), objects (data about learners) and goals (understand and optimize learning and environments).

Objects: Data About Learners

The rise of Learning Analytics comes from the chance of observing and tracking the learners' activities through log files. Logged data describes who the students are, which activities they carried out and when, and sometimes how and where, they worked. Such intensive data collection produces the so-called Big Data that facilitates the use of data analysis procedures.

Procedures: Measurement, Collection, Analysis and Reporting

Measurement and collection play a relevant role in the existing research. In particular, non-intrusive measurement and collection is difficult to achieve in the learning context. The most popular method is to capture web interactions in a Learning Management System, but the captured data may not be fully representative of the student activity, and other monitoring methods are required (Pardo & Kloos, 2011). Another main focus of these “methods” part relates to analysis and reporting, which are the more noticeable part of Learning Analytics because of the immediate usefulness. Methods include social network analysis, collaborative filtering, clustering, neural networks, just to mention some. The reporting methods put emphasis on visualization techniques, so that the sense making process is at the end done by humans judgment,and visual analytics (Keim et al., 2008) plays a relevant role.

Goal: Understand and Optimize Learning and Environments

Learning analytics attempts to discover the factors that affect learning in a certain context, so that instructors and learners and reflect on these factors and improve the teaching/learning experience. Tools can be viewed as decision support systems, where the decision is aimed at driving one's teaching/learning methods.

All in all, Learning Analytics is a research field that investigates methods for collecting data from a learning situation, algorithms to extract hidden information from the collected data, and knowledge representation techniques that have a possitive impact of the learning situation.

Key Terms in this Chapter

Data Mining: Field of knowledge aimed at the study of techniques to automatically analyse and extract meaningful information from large datasets. There is some overlaping between Data mining and Artificial Intelligences, and some techniques can be included in both research fields (e.g. cluster identification).

Artificial Intelligence: Field of knowledge devoted to provide computers with skills and abilities that are usually attributed to the human being. Some examples include pattern recognition, reasoning, natural language processing, learning, et cetera.

Dataset: A collection of data logged from a certain case of study. The later analysis of the dataset requires some metadata to describe the case of study, such as a description of the context, the existinig event types, the user profiles, et cetera.

Information Visualization: Field of knowledge aimed at the study of techniques to visually represent the information in a meaningful way, so that the representation have an added value over the raw data. For example, infographics are interconnected sequences of facts that have more information than a raw collection of facts.

Visual Analytics: A type of information analytics that relies the last step of the analysis (the sense making process) in the human being. This is achieved by the visual representation of the data that allows humans to easily recognise patterns and make conclusions that are difficult to achieve by computers.

Data Collection Method: A technique that observes the users activity and creates written reports of the observations. Methods are usually event-based, that is, each user action is reported as an individual entry shaped as ‘observed action, user identifier, timestamp’.

Big Data: An extremely large dataset that cannot be analysed by traditional methods, and require distributed, highly paralelizable computing to extract the information.

Complete Chapter List

Search this Book:
Reset