Learning Analytics: The Ultimate Guide

1. What is learning analytics?

2. How is learning analytics used?

3. Benefits of learning analytics?

4. Learning analytics data sources

5. Learning analytics methodologies

6. Learning analytics techniques

7. Learning analytics tools

8. Challenges of learning analytics

9. Examples of learning analytics

10. Final thoughts

We’re currently living in the Age of Big Data. And the advanced technology that ushered in this new age is only becoming more powerful. Never before have organisations been able to tap into such large amounts of information about their users, products, employees and operations. And this information has proven to be invaluable. In fact, data has even been referred to as the world’s most valuable resource. 

It seems that no organisation, sector, or business function has been untouched by the power of data – and this includes the training and education sectors. Due to the ability of data to transform and optimise learning experiences and outcomes, the field of learning analytics has been increasingly growing in importance. And as technology continues to evolve, harnessing the power of data will play an even larger role in enabling organisations to meet their learning objectives and maintain a competitive advantage.

In this comprehensive guide, we’ll cover everything you need to know about learning analytics and why it is so important to digital learning.

According to the Society for Learning Analytics Research (SoLAR), learning analytics is defined as the process of measuring, collecting, analysing and reporting on data related to learning. As a multidisciplinary field, learning analytics has roots in educational research, human centred design and analytics. Along with being an academic discipline, it also has practical applications within schools, higher education, educational institutes and corporate training.

The purpose of learning analytics is to harness the power to data to make better decisions about learning programs, strategies and environments. Using an evidence-based approach means that organisations are able to make decisions based on knowledge, as opposed to speculation – which often leads to better learning outcomes. 

Depending on the specific needs of the learners and organisation, learning analytics can be used in a myriad of ways. Below are some of the key applications of learning analytics:

Many learning analytics teams tap into the extensive learning data in order to better understand the behaviour of learners. As learning analytics can provide data visualisations on engagement and performance, they can understand the habits and motivations of their learners and make changes based on this information.

Another primary application of learning analytics is to gain the knowledge required to be able to effectively optimise courses and training. Learning analytics enable organisations to better personalise courses to the current learners, as well as use this information to be able to better improve future courses.

Many organisations use learning analytics as a way to better support learners as they are completing the course or training. For example, a number of universities use learning analytics to provide early intervention and support for students who are at risk of failing their course or program. By using learning analytics technology, instructors and advisors are able to identify and take action to help these students improve their performance early on.

The benefits of leveraging learning analytics for both organisation and learners are numerous – and will only increase as the technology advances. Below we cover the top four benefits that organisations can experience from leveraging learning analytics.

Being able to make data-informed decisions is one of the main advantages of learning analytics. This is because organisations no longer have to speculate on which areas of the course are effective. But rather, they have the data as evidence.

For example, let’s say an instructor believed that adding a short video to the module would help learners grasp a concept. However, through learning analytics, the instructor would be able to see that most students were not finishing the video until the end. The instructor could then conclude that the hypothesis was not correct and students didn’t actually find the video to be a beneficial learning resource. 

A further benefit of learning analytics is that it empowers organisations to use patterns and trends in the data to forecast future events and performance. This is known as predictive analytics and it is a powerful tool in helping organisations better understand learner behaviour.

One example of an application of predictive analytics in corporate training is using the data to make associations between the performance of new hires during digital onboarding assessments and later job performance. This information could then be used to improve the onboarding for future hires.

Since learning analytics provides organisations the data they need to improve digital learning more efficiently, it can lead to a reduction in cost. Without learning analytics, organisations would have to learn through trial and error – which would end up taking more time and using more resources. 

For example, if learner retention rates are low, learning analytics teams have more information at their disposal to more quickly determine what the cause could be. However, without analytics they would have to take shots in the dark – and may not be making the changes that actually lead to better retention rates.

Digital learning is typically a collaborative effort that involves a mix of various roles – including UX designers, training analysts, SMEs, instructors and learning coordinators. Therefore, effective collaboration and getting buy-in are key to success. 

However, with the different perspectives that come with the differing roles, it can be difficult for teams to reach alignment. And for this reason, learning analytics can be so valuable. It enables teams to get on the same page quicker since the proof is in the data.

Depending on the type of organisation, learners and learning outcomes, the data sets and sources that learning analytics teams use will vary. Below are some examples of common data sources within learning analytics:

• Login frequency: How often are learners logging into the course?

• Session metrics: How long are the learner sessions?

• Course progress: Are learners progressing well through the course or having challenges?

• Device used: Do learners prefer to log into the system using a computer or mobile device?

• Assessment results: Are learners struggling with any specific parts of the assessment?

• Feedback surveys: What does the quantitative learner data say?

• Participation: Are learners participating in the social learning activities?

• Resource usage: Are learners utilising all available resources?

• Completion rates: What rate of learners are completing the course?

The Society for Learning Analytics Research cites four common learning analytics methodologies that are commonly used. These are descriptive, diagnostic, predictive and prescriptive, which we dive into further below.

Descriptive analytics is a type of data analytics that describes what happened in the past. This type of data can help organisations understand participation and engagement rates, and whether  learners achieved the set goals. This type of analysis looks at the facts and figures to determine exactly what happened.

To understand why something happened, diagnostic analytics is often applied. Diagnostic analytics can help learning analytics teams spot patterns or anomalies in order to better understand the plausible reasons why something happened.

In order to make predictions about what is likely to occur in the future, learning analytics will typically employ predictive analytics. This type of learning analytics is helpful in predicting trends, and preparing for future outcomes.

When learning analytics teams want to make data-driven recommendations, they often rely on prescriptive analytics. This is a type of data analytics that is often automated through algorithms. For example, if a student hasn’t logged into the LMS in a certain number of days, they may receive an automated message.

Learning analytics involves employing a number of tools and techniques to discover patterns and relationships in the data that can lead to actionable conclusions. Two techniques that are commonly used in learning analytics are data mining and machine learning.

In order to make use of large volumes of data, organisations go through a process of data mining. This involves collecting, organising and filtering data in order to gain useful information. Learning Analysts would typically look for patterns or trends in the data from which they could make conclusions.

Machine learning is a branch of artificial intelligence (AI) that replicates the way humans learn from experience, and recognizes trends and patterns. Due to its ability to use historical data to accurately make future predictions, it has become a valuable tool within digital learning. 

In order to conduct learning analytics, organisations are reliant on an arsenal of tools that help them collect and sort the data. Therefore, choosing the most suitable and effective set of tools is a vital step in ensuring your learning analytics strategy is a success. 

There are various factors that come into play when determining which learning analytics tools you will need. Some aspects to consider are the sector you work in and which learning tools you already have at your disposal. Since many learning management systems include learning analytics features, this is something that should be considered.

Since some tools are specifically designed for education and some are specifically designed for corporate settings, you’ll want to make sure you’re selecting a tool that is appropriate to your learners and learning objectives. 

When it comes to choosing an appropriate tool for schools, universities and educational institutes, you’ll want to ensure the system is able to effectively identify at-risk students so that support and early intervention can be provided. 

Within education, it is typically the teachers, instructors and advisors who are responsible for monitoring student success. Therefore the tool should ideally include a user-friendly dashboard with data visualisations. 

Some examples of learning analytics tools that are designed for education are Intellischool Albitros and Blackboard Predict. There are also various Moodle plugins with learning analytics capabilities. 

When choosing learning analytics tools for corporate training or L&D, it’s crucial to align the tools and features with the business and learning objectives. For example, whether your focus is on employee onboarding, leadership development or upskilling, you may need specific types of learning analytics strategies and tools.

It is also important to ensure that the tool is user-friendly. While some L&D teams may have a specialised Learning Analyst, or Training Analyst within their team, this is not always the case. Therefore, choosing tools that enable you to make the most use of the data without advanced data analysis skills is also an important factor.

One example of an enterprise learning analytics tool is Learning Pool. Their reporting and analytics solution includes customised dashboards with data visualisations and AI-powered predictive analytics. A further example of a corporate learning platform with analytics features is  360learning, which uses AI to create personalised learning experiences.

Although there are numerous practical uses and benefits that come along with leveraging learning analytics, it is not without challenges. Below we address some of the main challenges associated with learning analytics and how organisations can overcome them.

Learning analytics often involves working with large amounts of data from varying sources. Some of the challenges that come along with working with large amounts of data include safely storing the data and effectively integrating it. These are two logistical challenges that organisations should keep in mind when planning their learning analytics strategy.

Although technology is able to collect large amounts of data, it is of little use without the skills required to understand and interpret the data in meaningful ways. Therefore, one challenge of learning analytics is attracting and retaining talent with the analytical skills required to effectively work with learning analytics.

Due to the fact that learning analytics deals with user data, ensuring data is properly handled and stored can be a challenge. It’s crucial that students consent to their data being collected and that the data is anonymised when possible. It’s also a good idea to provide employees with training on data protection in order to minimise the risk of data breaches.

Now that you’re familiarised with what learning analytics is and how it can be applied, we’ll have a look at some real-word examples of learning analytics in action. Below are three practical examples of successful learning analytics projects that can help you better understand the value of learning analytics.

One example of learning analytics being utilised in corporate L&D, is a project led by InterContinental Hotels Group (IHG). As a multinational organisation with thousands of employees, the objective of implementing the program was to improve internal communication in the workplace. 

With help from the learning platform Learning Pool, they delivered a Massive Open Online Course for thousands of employees. Following the course, IHG then used learning analytics to determine the impact of the course on the quality of online conversations. It was only through using learning analytics that IHG was able to understand how effective the course actually was in achieving the learning objective.

The University of Maryland Baltimore County (UMBC) used a predictive learning analytics campaign to contribute to better student achievement outcomes. The project involved configuring the analytics tool, Blackboard Predict, to send a “nudge” to students who may be at-risk based on their grade in the system.

The students received a message informing them that they were at risk of failing, and advising them of available resources. UMBC’s pilot was such a success that they subsequently rolled it out to further programs within the university.

A further example of the power of learning analytics within a university context is a project that the University of California Berkeley initiated to support struggling students. The goal of the project was to use learning analytics to determine early on in the semester which students may be having difficulties. This way, advisors would be able to intervene early and ensure students had the best opportunity for success.

The project involved integrating data from various sources, including their learning management system and their student information system, to create the Berkeley Online Advising system. The Berkeley Online Advising system enabled the university to scale their advising program and specifically identify students that were at-risk.

With the advancements in artificial intelligence and machine learning technology, learning analytics is a field that is continuously evolving. This evolution means that organisations are becoming increasingly more empowered to make evidence-based decisions that impact their learning strategies. 

Learning analytics is a powerful tool that has already transformed digital learning and will continue to do so. If you’re interested in learning more about learning analytics and digital learning, a Professional Diploma in Digital Learning Design is a great way to gain the foundational knowledge for a career in this exciting field.