Learning analytics in the e-Learning world

Lily Caroline

BI & Analytics

With the introduction of newer technologies into the IT world, there is an increasing demand of quickly learning these cutting edge technologies to be able to adapt it and be ready to apply it in practice. Books no doubt offer all the relevant knowledge that is required, but with the internet era having a strong foothold worldwide, e-learning practices have become a common practice and a preferred choice of learning.

What is e-Learning?

E-Learning is different from the conventional methods of learning. E-Learning is a new way of learning electronically. It leverages modern electronic technologies and online communication methods to provide access to courses to a wider audience worldwide. E-learning takes place either online, via CD/DVDs, or over TV channels. It is interactive where the mentor and the learners can have discussions and resolve queries. A true eLearning experience is the one that provides the right knowledge to its audiences and makes sure that every effort of the e-Learning program is driving them towards achieving their expected goals.

What does learning Analytics mean?

Learning analytics refers to the collation of data, performing analysis and reporting of data which is related to the learners/students, for the purpose of understanding the environments in which learning happens and optimising it accordingly.

Learning analytics refers to informative big data. Big data, makes use of standard statistical analysis methods to understand user experience trends. Today, it is inevitable for learning institutions to understand student experience, and customize educational courses as per individual student needs in order to predict student performance in their future planning efforts. Today, learning analytics are largely applied for student retention and student support.

The present State of Learning Analytics

• Adopting Learning Analytics isn’t very easy and may take a few years
• The presence of large amount of student data with education companies and universities may be at risk due to lack of adequate data security
• Wrong data inputs and inaccurate student profiling could cause potential issues for instructors while assessing individual student performance
• Learning Analytics, may not be able to create a strong impact on learning & education for professionals of the education industry and instructors
• In spite of all the odds, many popular education giants e.g. Pearson and others have slowly developed their own analytics system
• Big Data is relatively recent but has gained popularity and support by budding as well as experienced data scientists as well as professionals from the field of education

Key Practices for Learning Analytics

Learning analytics have established confirmations in the education industry, of being truly valuable tools. Any tool generally requires a minimum level of ability and understanding to be used to its maximum advantage and so does learning analytics.

Following are a few of the best practices in learning analytics arena:

• Define all data input requirements and constraints among stakeholders. Data must be consistent across the panel. Even minor differences or inaccurate data may cause skewed data
• Big data means pulling data from the Learning Management Systems and multiple other enterprise software systems used within organizations. It is the responsibility of all those involved with the system, and not only those dedicated to learning, to preserve integrity of data
• Data is not just names and numbers it can be much more than that. Many enterprises have the option to take in use-case studies as a part of datasets
• It is necessary to educate and impart training on how to use your reporting capabilities and tools, to all those who need reports on a daily basis. Many enterprises buy expensive Learning Management Systems with great capabilities but fail to leverage it to their maximum advantage to get best results

Learning analytics in an e-learning environment – Challenges faced

Learning analytics may not always work smoothly in an e-learning environment. Some common issues faced are as follows:

Lack of strong Collaboration: Learning analytics requires strong collaboration amongst, Mentors/instructors, IT professionals, administrators, and those involved in e-learning. At present there are few skilled and trained personnel who Mentor students. Collaboration between all the members helps to promote the creation of better results and helps in synchronizing various e-learning efforts. Collaborative effort from both, instructors and students/learners will result in a culture of leveraging data to enhance the overall e-Learning success.

The dispersed nature, wrong or incomplete learner data and its collation is difficult: The methods of education are changing and today learning online is open to all who are interested in learning various courses. Students make use of online resources, platforms and information networks. The dispersed nature of online student data or wrong/incomplete datasets makes data quality as well as collation of data a major challenge for learning analytics.

Limitations of Learning Analytic’s Tools: Mentors/instructors need to understand that the learning analytics software may not capture all student’s learning details. Although Learning Management Systems are a key source of big data, with the introduction of smartphones and mobile devices more and more students are accessing online courses through the smartphones/mobile devices. There is a need to understand that online courses and education should be considered beyond just Learning Management Systems. Research needs to be done to develop analytical tools that measure student behaviour as well as their experience and performance. Existing tools and learning management systems do not have enough capabilities to measure the entire student e-learning experience.

Need to identify factors that predict student’s social behaviour and responses, well in advance, before conducting analytics: E-learning Mentors/instructors should define measures and predictors of behaviors in advance, prior to performing analysis through learning analytics. When analyzing learning analytics data, it is important to consider human and social processes. Online student data requires interpretation by experts. More standardization is necessary for data analysis and interpretation.

What are the Implications for E-Learning?
Corporate companies leverage big data to make quick decisions. While learning analytics leverages online student data in an effort to change the way educational systems work today. Learning Analytics being relatively new there are multiple issues main one being, learning analytics not being completely adopted by educational institutions and practitioners.

Some important implications for e-learning are:
Many of the popular commercial enterprises, for example Amazon tracks customer’s online behaviour and related purchase activities and preferences. Based on their choices Amazon provides to them relevant personalized suggestions. Educational equivalences of such personalized experiences are under development in various educational universities and hopefully similar thoughts and ideas will be established soon in the e-learning arena in the years to come.

With the gradual increase in adoption rates of learning analytics, online instructors/mentors will have an added responsibility of developing their analytical skills in collating student data, interpreting it and providing valuable solutions to students. The mentors/instructors may need to develop another skill i.e. Data mining. They will also need to be proactive in learning the new analytical software tools and testing them for beneficial usage to the education system.

The learning analysis ultimately forms a structure or a framework, and the evaluation of learning tends to shift from the resultant evaluation to process evaluation. In-process assessment takes its input data from the day to day learning tasks and learning activities of students from within their social and informational networks and it forms the core of educational evaluation. With the growth of the data analytics field, embedded assessment techniques as well as decisive evaluation will become prominent in education. Not just embedded assessment, but interconnected feedback loops and systems will also be vital in forming and shaping up online education of the near future. Interconnected feedback refers to collating student data from online resources, analysing it and then interpreting it for sharing the ultimate data-driven decisions in the midst of involved stakeholders of learning universities and institutions.

Conclusion

In the e-Learning environment, learning analytics forms a systematic method of bringing together different techniques for collation of data, its analysis and visualisation of that data with a basis of teaching and learning process. Learning analytics renders ways for mentors and learners to enhance and develop themselves based on data, while they are either conducting or studying a particular online course respectively. The learning analytics results focus on important aspects such as supporting the development of vital skills: e.g. thought process, collaboration, applying ideas and writing clearly. Learning Analytics is beneficial to all parties involved in the e-Learning system.